Erwin Neher (2006) - Control of Neurotransmitter and Hormone Release by Calcium and Camp

It’s my great pleasure to speak to so many young researchers. Richard Ernst just told you what's important to get a job, that you have to know at least 4D, if not more. He also told you what's important to be happy, I’m going to tell you what's really important. And this is of course understanding what happens in our brain. We have known for about 100 years, due to the work of the famous Spanish neuroanatomist Ramon y Cajal and due to the work of other neuroanatomists around that time, that our brain is a network, very intricate network of billions of cells, which are shown here. This is just a picture of a few cells in the cortex as drawn by Cajal. Today, of course, we know much more about the propagation of signals in this network of cells. And particularly what happens in these cells. We have recognised that a neuron, like most other cells, have the basic ingredients of what the cell needs, which is the cell body, some organs, mitochondria and so on. But what's special about the neuron is that it has extensions. Some of them, or one of them, one of these extensions, the nerve axon, as you see here, or when you take this neuron, as you see here this long tube-like structure, along which an action potential, a nerve impulse propagates. And it’s also seen here that the neuron has another type of extensions, namely the dendrites which have these knob-like structures on them. And it is there where the nerve endings make contact, like here. Or let’s take this neuron again, this nerve ending makes contact with this neuron at a so called synapse. So synapses are the contact points between neurons. And what an individual neuron does is it accumulates, it integrates the information coming in on all these synapses which make contact on its dendrites, it integrates it, and whenever a threshold is reached of activity, a new action potential is generated and is propagated along the axon to then excite other cells. And it’s the enormous electric activity, which flows in this network, which accomplishes this task of information processing. And what neuroscientists have recognised in the last 30, 40 years, that it is the so called synaptic plasticity, which is at the very basis of information processing in the brain. Unlike in the computer where connections between elements are fixed in their strength. Synaptic strength, which is the signal mediated in this cell by an axon potential in the preceding cell, is not fixed, it is plastic in the sense that it constantly changes while information is flowing through the system. So just in short what happens at the synapse when a nerve action potential invades the nerve terminal, there is a 4-step process taking place. The action potential opens calcium channels, a special type of ion channels which let calcium ions enter the cytosol. The increase in intracellular calcium concentration indicated by this cloud of dots here triggers exocytosis, which means it triggers or it makes some of the storage granules, like this one here, which contain neurotransmitter to fuse with the plasma membrane and expel its contents. The neurotransmitter diffuses across the synaptic cleft and on the other side a neurotransmitter opens ion channels so that ion currents can flow into the postsynaptic membrane and create again an electrical signal. So what you have is a transformation of an electrical signal into a chemical one and back into an electrical signal. Now, what's synaptic plasticity? I said already that the term describes the effect that the signal strength transformed in such a synapse constantly changes. And, as I said, that neuroscientists believe that it is this constant rewiring, this constant change in the network properties which underlies some of the more complex information processing tasks of the central nervous system. This plasticity occurs on many time scales, on very short time scales, on long time scales, in the sense that if you stimulate a synapse a few times it changes its strength. If the stimulus is short, then this change is readily reversible. If you apply a certain type of signal over and over again, then the change that is induced by this is more persistent. And an important point of it, which I will come back later in a slide, is that nature uses all possible molecular mechanisms it has at its disposition, which were developed in evolution to achieve this finely tuned regulation of the synaptic transmission. And of course, when we study in the laboratories synaptic transmission today, we not only want to understand the mechanism per se, but, and this is the kind of problems that we address in our laboratory, what mechanisms underlie this plastic changes. So what do I mean by plasticity, and particularly by the short term plastic changes which we study in our laboratory? This shows more or less an overview over a few different forms of short term plasticity, which you can observe in different synapses in the brain. For instance, up here in this case a certain type of synapse is stimulated ten times, the first stimulus gives a very big response. The subsequent responses are smaller and smaller. If you plot response amplitude versus time, you see a decaying function. This type of plasticity is termed short-term depression. Another synapse, the same structure in the brain, gives exactly the opposite, a strongly facilitating response. A third type of synapse gives a combination of two, facilitation followed by depression. So each type of synapse which we can differentiate morphologically also has its personality in terms of the short term plasticity. Now, before going into detail of mechanisms, let me come to a more general viewpoint. We know that in simple organisms, and I could have shown here also an example from yeast, simple organisms, simple cells have a tremendously complicated network of signalling events going on inside themselves. And these regulatory networks to regulate metabolism, to regulate cell cycle, to regulate protein biosynthesis and so on, these have of course developed during evolution. And evolution took maybe about 80% of the available time to develop these very complicated networks which are basically in every cell. Now, evolution tends to reuse earlier inventions for new purposes. Chemical transmission is a relatively new invention of evolution. I mean chemical transmission became only important, of course, when there were multi-cell organisms and after a more extensive nervous system has been developed. So it is quite likely that evolution reused all these mechanisms to do this job. And indeed, probably the reason why nature does it so complicated to transmit a signal from one neuron to the other, the best reason for this is that it gave the possibility to reuse what it had invented earlier on. And I mean, there has been a long debate in neuroscience why is there so much chemical transmission as opposed to electrical transmission. You do also find electrical transmission in the nervous system, where there are simply electric connections between two cells, to pass on an electrical signal from one cell to the next one. But the majority of synapses, particularly those involved in the signal processing tasks do use this very complicated process of chemical transmission in order to reuse the inventions of evolution. So for this reason, and now I am coming back to my title, it is not surprising to find that calcium, one of the most important second messengers, and also cyclic AMP are among the regulators. And I mean, I don’t want to say with my title that cyclic AMP and calcium are in any way special, they are just examples, I could probably have taken any other example of an important signalling pathway in normal cell biology and would have found that this influences one or the other step in synaptic transmission. So in fact, due to lack of time, I will probably concentrate in my lecture only on calcium, give you one example about the importance of calcium in the real triggering mechanism and then just mention at the end at what places cyclic AMP comes in. So when we talk about the short-term synaptic plasticity, the fact that synaptic transmission changes, when you use this synapse repeatedly, of course you want to ask what changes. Now I explain to you the basic mechanism of synaptic transmission, release of neurotransmitter, opening of channels on the other side. So what I didn’t mention is of course, when a vesicle is used up, when it has lost its contents, it has to be reformed, the membrane which has become part of the outer membrane now, has to be taken up again. New vesicles have to be formed, these have to be filled with transmitter. They have to be transported to the right places and so on. So along this multi step process, you may have changes, and in fact you do have changes, while this short term plasticity happens. And from a mechanistic point of view, these changes can be changes in second messenger levels, they can be changes in phosphorylation of the proteins involved. There is, on the longer time scale, a protein synthesis, protein degradation, there are cyctoskeletal reorganisations who play a role here. And of course there are gross morphological changes like axon sprouting and formation of new synapses altogether. Now let me come to what we are doing in our lab. As I said before, most of the processes we study are just processes which are known in other contexts. Synaptic plasticity has been described as early as the ‘50s, you know. Nevertheless, a progress in understanding what happens has been difficult for two reasons. First of all, with this multi-step process where you usually can measure only – you stimulate a nerve and you measure the output, the result, the signal in the pulse synaptic cell. So it’s hard to associate certain change to a certain mechanism, and this has been particularly difficult because presynaptic terminals as a rule are very small, structures of only 1 or 2 micrometre diameter dimensions, which are not easy to access with electro-physiological tools. However, it has been known since the work of Cajal and even earlier, that in some places in our brain there are special synapses with very large terminals. This is a drawing, again from Cajal, who drew this kind of calices of Held very often because he liked them, he thought they were proof for his concept of the neuron, which at this time, 100 years ago, was a big controversy on whether there are neurons in the brain, individual units or whether the brain would be a synthesium. And he liked to use the calyx of Held as a weapon against the so called antineuronistas. Ok, so what is the calyx of Held? It’s a synapse in the auditory pathway. What you see here is a brain slice, so to speak a slice from the brain stem of a rat, where you see the signal path of an auditory signal. So sound creates action potentials in the ear, in the cochlea, which are transmitted to this part of the brain, transmitted through a synapse here in the ventral cochlear nucleus. And then the information and action potential crosses the mid line and makes a synapse onto another cell here. And the output of this synapse then again goes to another brain area, the so called lateral superior olive. And it is here in the lateral superior olive where the information of one ear first meets the information from the other ear. Now, we have a very complicated task to do in hearing, namely the sound localisation, directional hearing, in order for us to recognise, to about 5° precision where sound comes from, we have to know time differences between this information to an accuracy of about 20-50 microseconds. Also we have to very finely be able to recognise intensity differences. And for this reason, because information has to be transported so precisely, these two synapses are built in a very special way with very large presynaptic terminals like shown here. Presynaptic terminal surrounding a compact postsynaptic neuron like a chalice, a calix, this is why this is named calyx of Held. And in this very big synapse there are about 500 so called active zones, regions where you see accumulations of vesicles, where vesicles are released and meet a postsynaptic target. And the big structure here allows us to insert electrodes into both compartments and to measure electrical signals from both compartments simultaneously. We can apply all the so called voltage clamp technique which allows us to fix the voltages, the electrical signals here on both sides. And it also allows us to infuse through the pipettes substances into these two compartments. And what we use this for is to infuse into these compartments fluorescent dyes, like here. This pipette here contains the fluorescent dye which diffused into the calyx, and so what you see here in the fluorescence microscopy is the dye which has been distributing here. Also the dye diffused into the axon. And we use this fluorescent dyes to, for instance calcium indicator dyes, to measure calcium, and we use this also, as I will show in a minute, to infuse caged compounds. Now, the synapse shows synaptic plasticity, as I explained to you before, at a certain concentration of extracellular calcium, you see here the typical sequence of facilitation and depression when you stimulate the synapse with a 100Hz pulse train. And as I said already, we are interested to find out what is the basis of these short term forms of synaptic plasticity. In short, the facilitation which I mentioned is usually understood in terms of the build-up of the local calcium signal. When calcium channels open, they create so called microdomains of elevated calcium. The red colour here is supposed to indicate the concentration of calcium which flows into these channels. But, as this picture indicates, the domains of increased calcium concentration are very small when channels open only shortly. But when you stimulate a cell repetitively, if these channels open and close repetitively, then the calcium will build up and finally reach, after three or four action potentials, larger dimensions just will increase successively. On the other hand, the depression which ensues then is usually understood in the synapse as a depletion of available vesicles, in the sense that when you stimulate the synapse 1, a relatively large fraction of those vesicles which are sitting here ready to be used, will be used up, so that you have to bring to the membrane new vesicles, you have to regenerate the vesicles and this finally is the limiting factor in the synapse, which makes the signal decay as you keep stimulating. So, as I said I will concentrate on only one aspect of this whole thing, which is the role of this local calcium increase to trigger this process of fusion of a vesicle with the plasma membrane. Now, we have known since 50 years, since the work of Katz and Miledi and later Eccles, that a rise in this calcium is immediate trigger for what happens here. But we have not known how high calcium concentration has to rise at the point of calcium sensing because of this complicated geometrical arrangement, we know from calcium imaging studies that the increase in free calcium is very localised. But of course calcium imaging studies can not resolve to better than the wavelengths of the light, a few hundred nanometres. And the distances between channels and things are shorter than a few hundred nanometres. And depending on how close a channel is to the release side, you can have calcium concentrations in the range from 10 micromolar up to almost a millimolar. So now we made an effort to really find out what the sensitivity of the release apparatus is. And, as I said, this is complicated because we don’t know the exact distances, and so even from measuring calcium you cannot infer on what happens really locally. So help comes from chemistry. Namely we can apply to the interior, so called caged compounds, caged calcium, substances like EGTA, calcium chelators, which however are photolabile. So calcium DM-nitrophen is one of these substances which you can destroy by a flash of UV light to produce photoproducts and to release the calcium which is bound to it. We can do this readily in the clinics of health by infusing this caged compound into the presynaptic terminal, together with the indicator dye. So we can measure the free calcium concentration here by the fluorescence. And we can induce a jump in free calcium concentration by giving a flash of UV light. And this is shown here on this half. Here you see that in the first experiment we gave a weak flash which increased calcium to about 2 micromolar and you see that this elicited in the postsynaptic cell, in this pipette, a current signal which is very small and noisy. It is very small and noisy because it results from the release of only relatively few vesicles and we know, we can measure that a single vesicle containing transmitter, when it releases its transmitter, produces a small signal, a so called miniature excited or postsynaptic current in the order of magnitude of 20, 30 picoamp, which decays very rapidly. And what you see here is indeed the superposition of quite a number of such events, but still at a relatively low rate of maybe a few events per millisecond. Now, if we give a stronger flash and increase calcium to 4 micromolar, you can see that you induce a much larger current, this is a different scale now. Still it’s a little bit noisy because, of course, it also consists of a superposition of many of these elementary events. If you go to 6 micromolar, you induce a current of about 7/8 nanoamp, which is comparable to the current which you induce by stimulating the synapse in the usual physiological way, by an action potential. So a very first answer to our question, how high does calcium have to rise in order to produce a normal physiological response, would be you have to go up into the range of 6 micromolar. But of course this is not an exact answer because what we are doing here, we are increasing calcium and calcium stays up. Whereas in real life, when an action potential comes, calcium channels will open very shortly and will close again, so that you have only very short-lived, transient increase in calcium and you would expect that you would need higher peak values of the strengths of calcium to produce the same response here as compared to the case where calcium stays up. So we have to do more biophysical analysis and the kind of analysis we do here is not spectroscopy but deconvolution analysis. Deconvolution analysis again is a linear technique like spectroscopy, which assumes that the responses you measure here are the linear superposition of signals of this shape. Now, if you know the elementary signal, you can deconvolve these traces and get back as an output a time function which gives you the weight by which these individual signals occur and contribute to make up these traces. And as you can see here, when we deconvolve this trace here, the answer is that the weight of release rises very shortly after the flash to a value of about maybe 10 events per millisecond and then drops back again. For the bigger flash, the black curve here, we can see the rate rapidly rises to a peak value of about 60 release events per millisecond. It then drops down rapidly, in spite of the fact that calcium is still high, so this must mean that we are exhausting the supply of release ready vesicles and indeed the decay time constant here then would give us the mean time that it takes for a given vesicle to release at a calcium concentration of this amplitude. Ok, so we can do many of these experiments, analyse peak release rates, analyse the half rise times and so on, do a kinetic analysis on this. And we find, if we plot for instance the peak release rate versus the calcium value reached during the experiment, that this is a very steep function of calcium. The broken curve here has a slope of 4.2, which means that you probably need at least four calcium ions, if not five to trigger this process. We see that the delay shortens as we increase calcium. And we can put this data into a biophysical model where we assume that some calcium sensor x has to bind calcium, and once it has bound five calcium ions it can fuse. So the kinetic data allows us then to get the rate constants, the rate of calcium bindings, the rate of calcium association, the fusion rate for the quintuply bound sensor. And once we have basically characterised the kinetics of the system in this way, we now can go back to the question of what happens during an action potential, when we now just record postsynaptically, stimulate the afferent nerve, elicit an action potential in the afferent nerve and observe a postsynaptic current which has this shape. So we can deconvolve the postsynaptic current and get this time function which is the release rate, vesicles per millisecond, released during the action potential. We can see that the peak release rate is about 300 vesicles per millisecond, that the release happens only during a very short interval, only about the halfwidth of about half a millisecond. Now we can go into our biophysical model and ask which calcium concentration do we have to drive the model with, in order to obtain this black curve. And the answer is that we have to apply to the model a calcium time function as the red function, the red trace here. And it tells us that we infer, so to speak, that we need a peak calcium concentration of about 25 micromolar. That we need a calcium time course of this calcium signal of only about 400 microsecond halfwidths. If we give a broader calcium signal to the model, then the output will be too broad, the rise time of the EPSC would be too long. So from this we infer that the calcium transient has to be very short, peak of about 20 micromolar, only half a millisecond wide. Now, more recent measurements by a student in Bert Sakmann’s lab showed that actually if you induce experimentally by modifying the technique of caged calcium such short calcium transient, you actually have to apply exactly those wave forms which were predicted by this model in order to produce actionpotential-like responses. And biophysical modelling tells you that such high calcium concentrations are only obtained in microdomains around open channels. So this is kind of biophysical proof that indeed there is this very close association between channels and release sites. And that the signalling here happens very locally. Now, ok, I mean we studied many steps around this and of course got quite a number of more results. Let me just say a few words on the role of cyclic AMP also. Let me start from down there. As I showed to you, calcium is important for the triggering of exocytosis. In terms of molecules, this happens through so called SNARE proteins, specific molecules which are found in the presynaptic terminal, which are believed to form a complex to tie the vesicle and the membrane together. And then probably interact with a calcium binding protein by name of synaptotagmin in order to induce this rapid response. We looked at the role of cyclic AMP and found that this triggering step, this very fast process was immune to all kind of modulation by cyclic AMP and other things. So it seems that in this final triggering step, it’s only calcium which plays a role. However, in the preceding step of formation of this molecular machinery, in probably also the transport of a vesicle to the right place, there are two players. On the one hand there is another role of calcium acting on a much slower time scale, through calmodulin. And of the role of cyclic AMP, probably mediated through a cyclic AMP dependent nucleotide exchange factor. The two of them have to cooperate in order to obtain what's called priming, which means some final step, biochemical step before a vesicle becomes ready for release. Furthermore, in another series of studies we looked at the role of a kinase dependent process, and could show that phosphorylation of one of these so called SNARE proteins, the protein by name SNAP25, is actually important in this priming reaction. Again we found that phosphorylation of these molecules does not influence the triggering step but rather this priming step. And all of this work, of course, was not done by myself, but by young researchers in the lab, Takeshi Sakaba, a post doc from Japan, Ralf Schneggenburger, who is now a professor in Lucerne in Switzerland, Nobutake Hosoi, who is a postdoc in the laboratory now, Felix Felmy, who did his PhD study on this process. Thank you very much for your attention.

Erwin Neher (2006)

Control of Neurotransmitter and Hormone Release by Calcium and Camp

Erwin Neher (2006)

Control of Neurotransmitter and Hormone Release by Calcium and Camp

Abstract

Synaptic transmission is one of the main signaling mechanisms in the central nervous system. Synapses are the contact points between neurons and it is the signal flow across these junctions, which underlies the various information processing tasks, that our brain can fulfill. Unlike the digital computer, these connections among the computational elements, are not fixed in their strength. Rather, they undergo dynamic changes as they are used. In other words: The circuits of our brain are constantly rewired as information flows through the system. This phenomena is called ‘synaptic plasticity‘ and is believed to be essential for the mode of operation.

Synaptic transmission, i.e. the signal transfer from one neuron to the next, is a 2-step process. Electrical activity in the sending neuron releases a chemical message, the so-called neurotransmitter, which is stored in vesicles in the nerve terminal. The neurotransmitter then diffuses to the ‘receiving neuron‘, where it elicits electrical signal. It has been known since the work of Katz and collaborators in the early 50s, that an increase in intracellular Ca++ concentration ([Ca++]) in the ‘sending‘ nerve terminal is the immediate trigger for the release of the neurotransmitter. Later work has shown that next to Ca++ many other signaling pathways, particularly via cAMP, modulate the release of the neurotransmitter. These ‘modulatory‘ effects are central to the process of ‘synaptic plasticity‘. Similar molecular mechanisms are also responsible for the release of hormones from neuroendocrine cells.

Regulated release or secretion is a multistep process; the signaling pathways involved act at many stages and the question arises, which particular step is affected by one or the other of the common second messengers. Biochemical and traditional electrophysiological techniques very often cannot dissociate between signaling actions on ion channels, vesicle trafficking, and the secretory process itself. We have made an effort to dissect the stimulus secretion pathway by developing assays in chromaffin cells (for adrenaline release) and at a glutamatergic central nervous synapse (the Calyx of Held), which allow us to study secretion in single cells under voltage clamp conditions. This enables us to clearly distinguish between influences on electrical signaling from those on the processes of recruitment of vesicles and on the process of exocytosis. Our approach confirms that an increase in [Ca++] triggers neurotransmitter release and provides quantitative information about the spatial and temporal aspects of this Ca++-signal. Surprisingly, other modulatory signals, such as those mediated by cAMP do not influence this step, but rather enhance the ‘recruitment‘ of transmitter-filled vesicles, making them ready for release.

Cite


Specify width: px

Share

COPYRIGHT

Cite


Specify width: px

Share

COPYRIGHT


Related Content