 Okay. Good afternoon everyone. Thanks for attending to our new SAS exhibition talk. Today I have the pleasure of introducing our next speaker, Professor Enrique von Herzog from the Folium Institute. Thanks Enrique for accepting our invitation. It is a pleasure to have the opportunity of enjoying your recent research field. And briefly, I would like to give a brief introduction about Professor Enrique von Herzog. Professor Enrique von Herzog got his PhD on physics at the University of Minnesota in the United States. Then in the field of the neurobiology, he did two postdoctoral, two postdocs. Firstly, with Gary Matthews. And then he moved to Germany at the laboratory of Professor Erwin Meher in Gottingen for studying some really, really interesting mechanisms of vesicular pulmonary filings if I am okay with this. I remember that during my PhD, I read many of your paper, so you are sitting in many times on my thesis. So, and today Professor Enrique, so actually he's working at the Folium Institute in the United States and he's working as a full professor and senior scientist. So thanks Enrique again. So we are ready for enjoying your talk and thanks for accepting our invitation again. Thank you a lot Jose. It's a pleasure to be here. I've enjoyed a lot of these series of lectures, the Sussex Vision series. It's been wonderful to see the community. So I think it's one of the good things out of the pandemic that we, we found ways to still connect with each other. So thank you very much for the invitation. I hope you will enjoy the talk and let me start. I will hit my screen share here. Can you see this? Yeah. Good, you're, yep. Okay. So, so this is the title of my talk. And I'm going to be talking about crossover inhibition in the mouse retina, which is a feature associated with the A2 amicron cells. And this talk will also be concentrated on the glycine release from these A2 amicron cells, these specialized amicron cells which are narrow field glycine releasing amicron cells in the mammalian retina. The work is by Mark Meadows, Vera, he's a PhD student in our lab and Vera Balakrishnan and Shahan Wang contributed at initial stages of the work, especially the capacitance measurements. And so let me start out by, by showing you some pictures of what the mouse retina, since we're talking about the mouse retina has to worry about. And one of the things it has to worry about is these creatures here, this is the occipiter shark-shinned hawk, and it likes to hunt actually at dusk and dawn when maybe there's less contrast around. And in order for the mouse to detect this creature, you can already tell that it would be good if it had retinal ganglion cells that would be orientation selective so that it would be able to tell the orientation here of these two bars, let's say, dark bars. So orientation tuned retinal ganglion cells would be a good thing for it to have in its retina. Also directional selective, right? It wants to know if this bird is flying in this direction or that direction. And it also would like to tell the velocity, so the velocity at which the bird is flying in the air. And it has UVs, so there are UV cones that allow it to see well in the UV range. Here we have a lot of UV light during the day. And the other feature, of course, that it wants to see is this object getting smaller, which is a good thing, or is this object actually getting larger quickly, which is a bad thing for the mouse. So it wants to also detect whether this object is looming or not. And it has to do that in dusk and dawn when it starts to get hungry. And it's an octonal animal, the mouse, and it wants to start looking for food. It has to be having good vision. I think mouse actually have good vision. They cannot detect this just with their noses or with their whiskers or their ears. These are silent flying machines. You can see the aerodynamics of this. So the mouse probably has very good vision, a very sophisticated retina, to detect these objects. And these are really formidable creatures. You see the size of the eye is huge. You can think of it compared to the brain. And if you actually dissect this out, so I found this in the web, you can actually think of these occipital hawks as being flying, primarily a flying set of eyeballs. You see these huge eyeballs compared to the brain. So it probably has a lot of space here, neural space dedicated to vision to be able to scan and then swoop in and dive for its ability to survive through hunting. So contrast is important and contrast in low light conditions, dawn and dusk, when the mouse really needs to see these creatures if it wants to venture outside. So what are mechanisms that the retina has for contrast? This paper I wanted to quickly review. So in my initial part of my talk, I will review a few papers that I think are really interesting for this concept of cross inhibition. Which is how the on and off pathways of the retina can talk to each other. And this paper is really, really nice from Michael Freed, lab, talking about the efficiency of contrast encoding in the mammalian retina using this crossover inhibition. So if we think of the cones and how during dark, so if you give a dark flash here as your stimulus, the cone will release glutamate to darkness. It will activate these amper canate receptors here in the dendrites of the off bipolar cells, which will then release glutamate and cause an EPSC depolarization in these off alpha cells, these off ganglion cells, and that will cause spikes. So communicating to the brain that there's a flash of darkness in the field of view. At the same time, you will have also an inhibition so there will be glutamate onto these on bipolar cells, which will be hyper polarized. They're connected these on bipolar cells this on pathway here through gap junctions to these a to amicron cells, which is the main interest of my talk will be to try to understand a little bit better how these cells function. And these cells then receive that that hyper polarization from coming from the signaling from the cone, and that will inhibit these. The glycine release here the terminals of the eight to amicron cells, so that you'll get even more efficient release of the glutamate. And you might think that this and actually the release of the glycine occurs not only here pre synaptically at the terminal of the bipolar cell, but also onto the dendrites of the off alpha cell. And even the soma recently will Grimes and Jeff Diamond show that even the soma gets direct glycine release from these eight to so they're very versatile cells. And if you now block this pathway was a drug called LAP four. What happens is very interesting so if you don't have this crossover inhibition. That they show in the paper so in control you have a very low noise so you have here a little bit of of contrast 4% contrast and you get a deep polarizing current here. And, and here's another reproduction of this it's very highly reproducible low noise, very good signal, you put a lot before it starts to get very noisy, the baseline. And it becomes very difficult for you to distinguish what is signal from noise. And if you do an analysis of the noise and the signal here in control you can distinguish very well with these Fisher measures. But in LAP four you cannot distinguish anymore these low contrasts. So this circuitry here is vital for you to detect these low contrast features. And this is all in voltage clamp we have currents here. If we go to to to to current clamp here. And as we start to get spiking, we see that the important features for spiking are of course the resting memory potential of the off alpha ganglion cell, and also how distant that is to the threshold so here's the resting member potential the plateau potential where you start to get spikes you get a burst of spikes to signal to the brain, and where you have your threshold is very important also. So to depolarize more you get more of this plateau potential you'll you'll get more spikes. And you see that in control, you can be safely below the, the, the threshold for getting spikes. So you might not be able to detect very very low contrast, but if the contrast gets a little bit better you get a burst of spikes. In LAP four you get very noisy get spiking all the time because the cells now don't have that tonic inhibition, the resting their potential is more depolarized and now any little glitch here of the memory potential will reach the threshold and you get a lot of spikes, you might be able to get even some more contrast to some more signaling to very low contrast, but there's too much noise for you to really distinguish that. So, so you can see how important this a to American cell is because it sets the resting membrane potential at a sweet spot where it's not too hyper polarized where it's far away from the threshold. It's not depolarized where it's giving lots of spikes it's really setting very delicately this memory potential so that you get it very finely tuned to get very efficient contrast. So you can see how this mechanism of crossover inhibition can be very important for detecting these dark flashes. Another situation I was telling you about the hawk and the mouse and how if the hawk is looming the mouse wants to very quickly detect that send that signal to its from its retina as quickly as possible to its motor cortex so that it goes back to its burrow very quickly. So it doesn't want to have looming detectors or, or you know orientation bars in its v one it wants to have it in its retina to get the message very quickly to the brain. This group found is really interesting looming detector. And interestingly the a to is is critical here to in the circuitry. And they say that the circuits essential building block is rapid inhibitory pathway that comes in part from the a to so they studied these PV five ganglion cells that actually do not respond to lateral motion. They do respond to looming motion so an object that's that small and gradually gets bigger and bigger and bigger. They respond to that very very well with spiking. And in this paper they even did paired recordings between the a to and these p five ganglion cells. And it's really interesting, they, they could get pairs here is they show how the, these IPSEs are are blocked by strict nine. And here's the a to that they reconstructed here's the ganglion cell the PV five you see it has this large dendritic field. Here's one a to which are these narrow. These are the dendritic amicron cells. That's synapsing they show here these are the dendrites of the PV five. There's synapsing and here are the lobules that release the glycine onto the dendrites of these PV fives. So for looming detectors, maybe the eight twos are really a critical fast glycine inhibition that's important for these cells to distinguish from something that's moving lary from something that's moving. So really nice circuitry here for a feature detector. The last thing I wanted to say, sort of as a general overview of why a twos are so interesting. The, the highest density of amicron cells that we know is the a to so they're very important. This this paper I found really fascinating from Erica straight toy and Rania mastery and Rick Gruner, who's in Australia, and they looked and saw that they're a to amicron cells in the primate fovea and this is actually from human retina. This is the figure caption, and what they're doing here is this is just a beautiful picture. They're putting a rod arrest and to, to, to look at where the rods are and you see here's the fovea the foveal pit, and you see how the rods are peating out here, they're peering out, and you have only the cones and cone bipolar cells underneath here. They also image the rod bipolar cells so you see that the rod bipolar cells are peering out here, and they, they looked at call retinin as a marker for the a twos it turns out that it's a very, very good marker for a to amicron cells in primate retina and human. And here they even reconstructed in this this rectangle here is shown blown up here and you see reconstruction of these three little midget amicron cells which are really much going quite into the the fovea. So, pretty fascinating. They even in this paper speculate that a twos are probably very ancient amicron cells that developed together with the cones, and only later in evolution when the rods came in because you know mammals like, like, rats in mice, and they figured out that hawks sleep during the night. So they figured out we better become nocturnal so we don't have to face those hawks. And so they started to develop rods to for night vision and co opted probably the circuitry was the a to so that it would serve the signaling of single photons, for example, the low noise pathway for single photons. So the a twos are pretty amazing they're they're involved in in the low noise single photon circuitry but also in the daytime. And this as this shows very nicely. So a twos are ancient cells that probably co evolved with the cones for this crossover inhibition their multitaskers as I was showing you, they, they are involved in detecting lots of features not only in the circuitry but in the photons, where it's scotopic, but also in the photopic range. And so they have a rich, let's say social life because they're connecting receiving inputs from rod bipolar cells and cone bipolar cells, and they're also hooking up with many different types of bipolar cells sending, you know, junctions electrical sinepses chemical sinepses to different bipolar cells and and also releasing glycine to retinal ganglion cells and setting their resting memory potential in a very finely tuned way. But the question I want to ask today is, Okay, they have this nice social life but what about their interior life. I don't mean that in sort of, you know, a spiritual way. I mean it more in a biophysical mechanisms way. So what are the, the calcium stores and intracellular messengers. If these cells are such multi good multitaskers they probably have a rich internal life of lots of different second messengers and G protein coupled receptors and intracellular messengers like cyclic AMP and phosphate and calcium to signal. And actually we don't know too much about how second messengers are modulating the release of glycine. And that's what I want to tell you about how our lab is trying to approach that. And, and see how this rich interior life can facilitate and promote the multitasking the a twos. Okay, so this is just a review here of the retina. These multiple parallel pathways, lots of different types of bipolar cells. And I want to focus on this glycine release that's very important for crossover inhibition, so that these excitatory signals coming in here are converted in the a to American cell to inhibition in the off pathway. So these a twos are central to both scotopic and for topic vision. We used to think, you know, when I was starting graduate school that it was just specialized for for scotopic vision. But then we started to see was this disinhibition and contrast enhancement and the looming detector that the a to not only has, you know, night shift is Jeff diamond has famously said together was Nick Oish, but also has a daytime job, a daytime gig where it it helps to detect all these features during the day. So here we're going to look at these dendritic inhibitory release sites at the lobby appendages. So these are called lobular appendages over here, these little sort of bubbles connected by next here on on the American cell coming out from this thick apical dendrite. We also have gap junctions. So quite complex for very small cells, doing a lot of different tasks in the retina so pretty interesting. And the way we've looked at the glycine releases was use this technique, where we can look at living cells and measure their membrane capacitance to look at the recycling of synaptic vesicles the exo endocytosis of synaptic vesicles in synaptic terminals. And we, we, I learned the technique from Gary Matthews when I was a postdoc in his lab. And we've he had this wonderful preparation which was the goldfish giant mixed bipolar cell terminals. And so here you see the goldfish bipolar cell, very similar cells but much smaller in the zebra fish retina. And actually acutely dissociate these cells from the retina they often come with their axon cut. So you have this perfect sphere that's ideal to measure membrane capacitance which is a measure of the surface area of the cell. So we would measure the capacitance and then depolarize it elicit an L type calcium current get calcium influx which triggers the exocytosis of vesicles, which triggers the increase in the surface area or the capacitance jumps that you see here. And then the vesicles after they fuse and increase that surface area, we found that they were re internalized so the capacitance return quickly back to baseline. So we interpreted this return as being endocytosis and so here is just a summary of the work that Gary Matthews has done. This all comes from the lab of Gary Matthews, that there was very fast endocytosis endocytosis occurs immediately was in 100 milliseconds you already start to see a decay of the capacitance so very very fast. Endocytosis was first found in these nerve terminals. We also went on to study by changing the pulse duration the, the release of the, the nerve terminals. And we saw that there was a very fast phase so if you take the derivative first derivative the capacitance you see that there's this spike a very fast release phase followed by a slower more delayed or or tonic release. And this turns out to look very much like what the glutamate that's being released from the nerve terminals looks like there's a fast component in a more delayed sustained component different bipolar cells will have different kinetics for different temporal frequencies that they're encoding. Some will be very fast and transient and others will be more sustained. And this all comes from the work of Gary Matthews. He was just a wonderful person and a wonderful mentor for me. And when I was transitioning from physics. He really taught me how to become a bio physicist, even more so a biologist to think like a biologist in a quantitative way, and had a lot of contributions to the retina and synaptic transmission. He was a biologist. Just passed away way too young. But we use then this technique in by recording from the soma's of the bipolar of the a to amicron cells. And our hope was that since these necks here are very short and these ends here are sealed their small little bubbles. There's their short here you see them here. That this would be electronically quite tight at least this part here, and that we could still use the capacitance so although this is by no means a sphere which is ideal for capacitance, we could still clamp this area here very very well. And so we just tried to see if we could transport this technique of capacitance measurements to these a to amicron cells. And in fact it worked quite well. If you just hold the cells at minus 80 millivolts and depolarize. You see, and you put a sine wave here of two kilohertz, you see that that depolarization elicits a calcium current, these cells have a single type of calcium current in that way they're simple. And that causes this jump in the capacitance. There is a change also in the series resistance we typically see a small downward shift is you wouldn't predict exactly. Also happens for example in the mossy fiber terminals. Stefan Hellerman has seen this in his capacitance measurements. These looked topologically very much like a mossy fiber terminal was the release sites very close to where the patch pipette is. So it seems like it's okay to measure capacitance so we went ahead and did that and published already several years ago a paper on that. And you see that as you increase the pulse duration you get bigger jumps more exocytosis. The exocytosis doesn't wash out for the first 10, 15 minutes. It's pretty stable. And so you can now do the same kind of experiments that we were doing in the bipolar cell with Gary and see that there's a very fast initial release. So also very phasic release and subsequent paired recordings by Cole Grandin and Jeff Diamond and and Meg Furukhi and Espen Hartweit between the A2 and the cone bipolar have shown that they have a very fast component of glycine release followed by a more sustained component of glycine release very much like what this capacitance seems to suggest. Okay, so we can do capacitance. And how does the ultra structure look like? Are there synaptic vesicles? And yes, this has been known. There was this beautiful paper from Strettoi, the same one who was working on the phobia. I should have showed you before. She worked with raviola in the show in the early 90s, mid 90s, and had these beautiful electron micrographs. So the A2 amocrine cells have these beautiful active zones. They're packed with vesicles. Here you see three different active zones, even very beautiful clathrin coated vesicles here and even some clathrin right at membrane. This is the cone bipolar cell terminal here, receiving the glycine. And here's another amocrine cell that's synapsing also here onto the cone bipolar cell terminal. So a lot of microcircuitry here processing the signal. This was a ultra structure from the rabbit A2 to off cone bipolar cell synapse. So multiple active zones, which is probably what's giving us that big capacitance jump. Large, actually the capacitance jump was 40 femto ferrets, which is equivalent if a single vesicle is 48 to ferrets to about 1000 vesicles. Quite a large number of vesicles. That's over the whole surface of the cell, of course. And so there's a large reserve pools because it just seems to not really deplete. If you be polarized not too strongly, you can keep on releasing. So maybe there's a recruitment process of vesicles here towards the active zone. And we asked questions about modulation of G proteins and dopamine and cyclic AMP. Dopamine seems to act through cyclic AMP. There's good evidence that glycine can be modulated by dopamine from the lab of Erica Eggers. And we're also interested in the stores. So what's the interior life of these cells like So the first thing we looked at was cyclic AMP and that's derived from ATP. So we did an experiment where we just simply put one millimolar cyclic AMP into the patch pipette. And we record the capacitance one minute, let the cycle can be dialyze fully into the cell. And when we did a second measurement, we saw a nice potential. And the calculation does not happen if we don't put one millimolar cyclic AMP. And so here's the average data. We see a highly significant increase, a lot of variability. So one A2 to another has very different release. But on average, there was a highly significant increase whenever we had one millimolar cyclic AMP in the patch pipette. The calcium currents did not change significantly. So this is the total charge in the calcium currents. We then went on to look at the mechanism and the mechanism is very interesting. It turns out that the cyclic AMP is not acting through PKA, which was our first guess. Actually PKA acts at the gap junctions in the arboreal dendrites of the A2. But for the glycine release, it's this molecule called EPAC, which is not a kinase. It binds to cyclic AMP, has a large conformational change, and then can act on different targets. And so to show that it was EPAC2, we used a selective inhibitor. And was that inhibitor and one millimolar cyclic AMP, it completely blocked the potentiation without significantly changing the calcium currents. And whereas if we use the selective acyclic AMP analog, so something that selectively activates EPAC, we saw the nice potentiation still occur. So this was very good evidence that EPAC is involved in this story. And in our paper, which was published back in November in journal neuroscience, we show that there's an increase in the vesicle pool size. And this does not, without any change in release probability, we did that with her pulse experiments. We put a blocker called H89 of PKA. That does not block inhibition. So we don't think PKA is involved. I'll show you some experiments where we use foreskilling to directly activate adenylate cyclase to increase cyclic AMP in the cells. And that actually works beautifully. So more evidence that there's cyclic AMP through adenylate cyclase. And then I'll show you some evidence that there are internal stores involved here. So let's get at how we looked at foreskilling. And so for that, we wanted to do a different set of experiments where we would patch the off bipolar cell and look at how glycine, spontaneous glycine, miniature IPSCs are occurring just by patching these off bipolar cells. And we initially did it in rats because these off cone bipolar cells are very small, hard to record from, and in rat they're bigger, so it's a little bit easier. We also did it in a transgenic mouse, a CIT2 transgenic mouse. I'll show you some of that just in a bit. And we can distinguish very well the off cones from the on cones, which are shown here. And here's our data. So beautiful signal to noise ratio. You see that before treatment with drugs, here's the noise, here's the signal, very nice minis occurring spontaneously, completely blocked by strict nine. And if we put foreskilling, we get a nice potentiation of the frequency of the events. And the analysis is shown here, we had a large number of events. And you see this is the inter event interval. So we're getting more short intervals, which means higher frequency of these events. And that's what's shown here. So here's the semi log plot showing better these small intervals. And here's the frequency plot showing that on average, there was a significant increase in the frequency of the minis. It's about more than a 50 fold change. We also looked at the amplitude and the kinetics of these minis, they're very fast, surprisingly fast, so about two milliseconds to decay, whether we had it in control or with foreskilling didn't really change significantly. Here's the amplitudes so nice skewed distributions. And the amplitudes didn't change very much with foreskilling. This is the transgenic animal we we use this was actually a trick we we we heard from Cole Grandin and Jeff time and they use this in impaired recordings. And it nicely labels the the off cone bipolar cell for certain class of off cone bipolar cells that receives this class energy input. Here's a typical experiment one experiment we did where we put some foreskilling. And you see there's a baseline rate of minis certain frequency of minis then after just a few minutes it starts to rise and stays up. And so here's average data you see after a few minutes here for five minutes, you start to get significant increases in the frequency, and it remains highly significant later on, after eight or nine minutes. Okay, so this collaborates what we're seeing was the capacitance. Here I just wanted to show you some nice images of that we patch these cells was the sit to a nice spinning cells so we, I like this video because it's just beautiful how the cells have this intricate nerve terminal. And this is from Heinz Wessler's review. And so this is the type of off bipolar cell that we're looking at here that has this sit to expression. Okay. And then just for comparison I find this interesting to compare these glycinergic currents in the terminals of the off cone bipolar cells with the glycine remember I was telling you that the glycine. It also hits the retinoganglion cells and in both the dendrites and also even the soma now from the recent work of Will Grimes and Jeff. And here the decay is much slower. Here you see the time constant is 20 milliseconds so actually if you were to fit these EPSCs average them out it's more like 20 milliseconds decay here. Very different kinetics, and I just love this picture from my old friend Dario Prati. And in the lab of Isabelle Liano. We overlap this postdocs when in gutting in. And I was just amazed by these records where you see the decay of the EPSC and then single channel currents occurring here. And you can see that the distribution of these IPSCs in the retinoganglion cells quite nicely Gaussian, big amplitudes and large conductance for these glycinergic ion channels. So interesting comparisons. Okay, so the next thing we wanted to look at was stores, we did some pharmacology. There's two drugs that we used one is called two APB we got this from paper of Martin Wilson where he did is also the dendrites of culture chicken amicron cells and he. This drug and it's claimed to be an inhibitor of IP three it's actually kind of dirty drug, it might be a better inhibitor of these. So see the, the store operated calcium exchange channels, and it might be depleting the calcium stores so it's disrupting the calcium stores so we tried it. And in fact it did block the potentialization with cyclic AMP. Here, we're put leaving the two APB 10 minutes on also the round it in for 10 minutes and then we're doing the wholesale patching so letting it incubate into the slice. So we blocked the potentialization cyclic AMP, but there was also a slight drop of the, the capacitance, the, sorry, the calcium current. So we did right on a dean we did two different concentrations of right on a dean. And here you see that there was a very significant very nice block of the potentialization was right on a dean. And overall very small change in the in the custom current. Actually, if we use 10 micro molar we saw no change in the calcium current and a complete block here of the potentialization. And this was encouraging we think it collaborates very well, British finance for Martin Wilson, and Chavez and diamond also had nice nice papers, always the rod bipolar cells showing that they also have calcium stores. But, but they were careful in their paper to say that this is not coming from the a to a micron cells some other glycinergic cell is actually doing this. Okay, so there seems to be stores that school. If there's stores, then it would be very interesting to try and experiment where we put a high GTA to see if we can block that effect. That's what we did here. We first did control experiments was just to a GTA which is our standard we see beautiful potentialization here. And then in the same slice we change the patch pipette and use 10 GTA. This is now unpublished data that I'll be showing you next, and you see very nice. This block was 10 millimolar e GTA was still one millimolar of a cycle MP in the patch pipette. So this collaborates very nicely this idea that there are stores you can block the calcium from the stores which are probably a little bit distant from the active zones. Interestingly, very interestingly, the total capacitance jump didn't change when we went from two millimolar GTA to 10 millimolar GTA. So this seems to suggest that there are nanodomains in in these cells. So one conclusion we can have is that these a twos have a tight nanodomain coupling between calcium channels and the exocytosis calcium sensor whatever that is some type of synaptic tagman probably. And the release from the stores that the release of calcium from the stores is further away from prime vesicles because we can actually block it with 10 millimolar GTA. So calcium stores may mobilize vesicles. Our idea is that in control situation, the cycle MP is releasing the through E pack this calcium from the stores. And that mobilizes vesicles that might be in the cytoplasm so that they move to the active zone and increase the red release pool size. We would like to get some more evidence for the stores more direct evidence. So we use the Glidey to cream house to try to express some GFP. So first, Glidey to was a bit controversial whether it's really expressed in the eight twos. So we, we did a cross between this Glidey to creep and the flock steady tomato and we saw beautiful fluorescence all over the retina. This is actually from Ben severe, beautiful images that he obtained was a confocal here you see in the INL the cell bodies and here you see these little lobular appendages in the IPL. So really looks like the eight twos. Here's a cross section. It's a bona fide a to sell that this Glidey to is is is expressing the teeny tomato on. So then we crossed with G camp, you know, the calcium sensor to nightly encoded calcium sensor. And now we can do experiments for example like this one, where we just patch the a to and depolarize it and see what happens. And we see really nice signals in the lobules. As soon as we depolarize the cell so the lobules just light up and the soma also light up lights up. So let me stop this video and so we did analysis of the different lobules and you see very nice rising fluorescence and then a decay. It's kind of cool to compare the, the kinetics, the lobules decay much more quickly than the soma. It's very similar to a nice study that Tom and Leon did on the terminals of the bipolar cells and smaller terminals have very fast decay, because the surface to volume ratio. Being so favorable and here we compared the fast time constant slow time constants, and the lobules have just a half second here time constant. Whereas the somas have this slower time constant. They decay more slowly. So there's the analysis of the imaging. Now we could do more interesting experiments, which is to put a cycle game piece. So there's this drug eight eight PCT. That's an analogous cycle MP that goes right through the membrane so we can just puff some did some quick and dirty experiments was just puffed some of this drug onto the retina and there you see it lighting up the a two, and you're going to see another to their flickering in here. And there you go. So it's like lightning bolts in the sky here the flickering of these eight twos whenever there's cyclic amp releasing calcium from stores we hypothesize. Here's just analysis so we looked at these two regions here. And then after a few seconds one a two lit up few more seconds another a two lit up when we're perfusing the this drug over the slice. And here we looked at different regions of interest. Here's the analysis of the data so we put the eight CPT and here's little bursts of calcium coming out. So this is also unpublished and kind of preliminary data which we want to analyze further. We repeated this here. Some signals are more sustained some are faster. They occur quite quickly once we put the drug in. We tried caffeine, same cells. So also saw caffeine, and this cell here I think had way too much caffeine and just petered out probably zonked out on the caffeine and is not recovering. So okay, let me reach my conclusions then. I showed you that there is a rich interior life in these eight twos that they have a dental like cyclase that can convert ATP to cyclic amp probably through some G protein receptor we're not sure what that is maybe dopamine, activating this E pack molecule which creates these ryanidine IP three channels releases calcium promotes the recruitment of the vesicles to the active zone, where they lie was in nanometers of calcium channels for fast release remember that looming was critical that inhibition very very fast and the eight twos provide very fast inhibition. So we think there's a dental like cyclase through E pack and calcium stores interesting mechanism that increase the pool size what are the unknowns Well we don't really know what the GPCR the the G protein coupled receptor signaling is what is is it dopamine is it some other neurotransmitter we're we're working on that. We also think there could be maybe E pack working on rim and directly on the fusion machine or there's evidence from insulin cells that that might be happening. So we want to look at that. And then there's some interesting mobilization of vesicles maybe through proteins like piccolo to the active zone to to get this pool size to be increased. So those are some of the questions we want to explore in the future. The main person who did this work was Mark Meadows really talented wonderful student that I've had in here he is actually in Norway he visited make for rookie and it's been hard fight in Norway I think this is one of the the tallest peaks there in near Bergen and these are collaborators at OHSU Ben Severe Catherine Morgan's Anusha Mishra Steve Smith. All the people in our lab, also a shout out to Teresa Putzeri and Roland Taylor who helped us a lot in these a two recordings. And make for rookie who hosted Mark. He learned a lot with them in a very short stay there in their lab, our funding agencies, and here is dusk in in Oregon, and you can actually see Hawks flying at that time of the day. And here's Mark in Berlin, and he's gotten a taste for beer. And I think he's probably going to do a postdoc in Europe. So I thank you for your attention. And I'm ready for questions. Okay, and make it. Thank you very much for such a beautiful talk. And yes, actually, we are looking for the questions of the people. So I have a question. So I have, I will take advantage as a host. And one of the particular one of the key aspects, I got surprised is that under your manipulation of the calcium stores. You didn't see a significant difference on the endocytotic decay on your capacitance traces. Yes, I would like to ask if you think that on this micro in sales. There is any effect of the modulation by the release of internal calcium stores on the endocytotic decay, because in other models. It seems to be related, like in the, you know, IPA campus, or even chromatin cells. Yes, yes. Yeah, yeah, chromatin cells are really interesting behavior was fast endocytosis. So in this study we haven't looked at the endocytosis actually endocytosis is very difficult to study, because at least in the cells we're looking at which were postnatal day 30, you know, so fairly mature but not really more mature. There was quite significant rundown of the endocytosis it occurs at the beginning after wholesale and then it runs down in more mature cells we saw better, more resistant robust endocytosis. So but we haven't explored yet if cyclic MP effects that kinetics with something we're studying now very actively we're very interested in the endocytosis the mechanisms that of these cells. Okay, okay. Yeah. So, Leon is asking is saying the results manipulating CNP and calcium signaling nicely show the potential for modulation of listening, listening energy inhibition. What do we know about presynaptic receptors on a choose that might accept such effect. Yes. So what are the, what is the signaling so dopamine is probably one of them that can affect cyclic MP levels and there's really nice papers by Erica Eggers in Arizona and she shows with light stimulation that the glycine is actually inhibited with dopamine. And so if dopamine is increasing cyclic MP one mechanism that could potentially explain that is we also find in these cells that there's calcium activated potassium channels. If the stores are releasing calcium they could activate these calcium activated potassium channels and that could inhibit the release. So there could be very interesting calcium modulation that inhibits release but also enhances release, depending on how your neuro modulator could be completely different at night as you know with melatonin and other things at night. I think there could be quite a bit of plasticity, maybe not as dramatic as yours Jose where the zebra fish pinch go blind right at one or two at night and then and then their ribbons kind of disappear and then they reappear at six in the morning. Maybe not that drastic in the mouse. But probably something similar, you know since dopamine levels change was light. It could be something completely different from dopamine that's actually elevating the the glycine probably multiple modulations yeah. So, Tom is asking the calcium, the calcium videos seem to suggest highly synchronized activity across the whole a two, which perhaps hints that the main driver for calcium is voltage. Now, do you see sign ups like or independently. So that's a very good question it could very well be that there is a depolarization. And that then causes calcium influx which then activates the stores so for example Ryan Dean receptors are activated by calcium going inside the cell. So, and in that imaging, of course we're not controlling the membrane voltage so we don't know. So those are future experiments that we need to do is actually do current clamp for example why we're putting those drugs and see if there are voltage dependent changes. If it's all just internal and the member potential doesn't change or the voltage dependent chains completely agree with Tom it there's probably a bit of both, since there's probably global and local changes. Yeah. Okay, so chat grabner is asking if can the calcium stores directly support release of glycine. I'm nice talk. Sorry. Sorry, sorry. If can the calcium stores directly support the release of glycine. Yeah, so that's an interesting experiment so we could, for example, block the calcium channels be a little bit artificial, but but it would be an interesting biophysical experiment we could block the calcium channels, and then put our impact activator release calcium and see if that's enough. My bet is, is maybe not too much. Because we could block it was the tenant GTA, indicating that the stores are a little bit far away. And so blocking with with the calcium channels might affect that. But that's a that's a good idea and it's an interesting experiment. Okay. Yeah, at the moment, there is not more question in the chat. I have another question but maybe something that question. Yeah, because you, I think I missed the value, but you mentioned the size of single vesicles with synergic single vesicles. Do you have any idea about the size of basic vesicles of people ourselves and how they relate. If they are so you decide some glutamate vesicles are bigger on people ourselves, or I'm a cream. I don't know it's just. Yeah, that's a very good question. There's a paper I remember from Tom Reese where he compared with EM, the size of of of a GABA ergic glycinergic vesicles maybe even and glue metergic, you know, glue metergic or somewhere around. Yeah, about 3540 nanometers. Yeah, it's around GABA ergic vesicles in EM tend to be a little bit flattened might be an artifact of the fixation. They was rapid freezing they tend to be more spherical, but they're about the same size, they're pretty uniform, which which is from the clathrin coat, you know, the size of the vesicles kind of dictated by the clathrin coat. So if you think the vesicles are budding off for example from endosomes after the endocytosis that, or maybe directly from the plasma membrane. They come already was kind of a uniform size so I think they're pretty similar, pretty pretty much around this value of 40 ATO ferrets. We have another question. Simon Laughlin is asking if is there any evidence that the signups that act presynaptically within the synaptic complex are regulating basically released by their posts synaptic partners. Yeah, that's an increasing question so I guess Simon you're talking about retrograde messengers maybe right. If I understand that correctly whether the post synaptic partner could signal retrograde leave. We have not seen like retrograde synapse that way. One interesting thing we did publish a few years ago in the Goldfish retina where we did paired recordings was that if we stimulate very strongly in the Goldfish retina the the MB by Polar Sal terminals those big we see that in about 30% of the cells. There's a really interesting potentiation of the amp current. So the glutamate does not change you can just stimulate that repeatedly and it doesn't change the capacitance doesn't change over a period of, you know 10 15 minutes, but the these get bigger and bigger and bigger in a percentage of amicron cells. And we figured out that the mechanism was calcium permeable ampereceptors. And we think that they are postsynaptically an incorporation of more ampereceptors maybe a phosphorylation of the ampereceptors that is causing a maybe a more long term plasticity in the retina. So it's kind of a taboo in the retina that to talk about LTP you know and LTD, but there could be post tetanic potentiation there could be you know after effects. When you go into dark room and you stay for a very long time in there and you go back to the light you know it takes you several minutes to readjust. So the retina could have after effects that take many, many minutes to readapt and could involve some postsynaptic plasticity and some retrograde messengers. Yeah, that's a great, great question Simon. We don't know what the retrograde messengers are we know that there's CB one there's endocannabinoids in the retina. There's pretty much any kind of you know there's no that might be regulating the vessels. There's there's a lot of retrograde messengers in the retina. Okay. We have another question. So any roles for PMCA and mitochondria in calcium handling. Yeah, that's a really good question we did not look at mitochondria. And that's something we looked at in in hair cells. So if you're very interested in their energetics, you would think that the a two, it does have quite a bit of mitochondria was all this activity this tonic inhibition that it's giving to set the resting memory potential of the retinal ganglion cells the off alpha retinal ganglion cells that it's, it's spending a lot of energy with all this inhibition that it's continuously sending to tone that resting memory potential of the ganglion cell. And mitochondria probably very important. They are a sync of calcium so when there's calcium around they ramp up their ATP production so they want to sense calcium, of course, but we haven't studied them per se. Okay. Okay, so I think we are finished with the questions of the live chat. And we, we already post the zoom link if you want to join us for the discussion now in more detail with Enrique. We already post the zoom link so feel free to, to join us. And thanks Enrique for accepting our invitation. It was a pleasure. And yes, it was a pleasure. So do I just stay on Jose. Yeah, you could just stay on so we will, we will, we will end the streaming in a couple of minutes. So we will give time to people for joining us. Oh, last question. Yeah. So, can synaptic input from raw BIPOLAR cells versus gap junction input from con BIPOLAR cells lead to different outcomes in calcium influx and this and glycine release. Yes, certainly, certainly. One thing that we didn't show that that we actually showed in a paper where we were looking at that plasticity from a calcium purple amper receptors is if you just puff amper onto the lobules of these A2 amocrine cells, it's really interesting you get calcium rises. And we think that the lobules have calcium purple amper receptors also. Not only their Boreal dendrites as you know Chavez and Jeff showed, but also the lobules. So if you just pop calcium channels and MDA and just puff amper, you get these calcium purple amper receptors and it could be that that calcium influx is enough to cause some glycine release and that's input that's coming from the off bipolar cells. So, so it could be that they are depolarizing and directly, you know, causing calcium influx that causes that the glycine, which is a very different mechanism from what the rod by polar cell does which it was amper receptors that go from the Boreal dendrites and dendrites that need to then depolarize the whole cell until there's a release of glycine. So there could be a differential modulation in in that way from rod bipolar cells which are on type and the off bipolar cells that are connected to the Boreal dendrites also via calcium purple amper receptors. So, it's interesting the calcium purple amper receptors appear a lot in the mammalian retina. But they're a feature more of very young developing synapses in the brain, but they remain in the adult retina so it's kind of an interesting thing maybe not explored enough. So, of course, a plasticity I think for the retina also. Okay. Okay, so thanks everyone for attending. So we will move now to the discussion and okay and see you next week. Should I stop sharing. Yes, yes, okay. Yes, maybe I can stay here and I don't know, can you guys see these these slides. Yeah. Yeah, yeah, we can. Yes, I can. This is a beautiful paper. I know if you guys saw from Espen hard fight and make for rookie and and Zant, where they do triple recordings also on the eight twos. And, and they show how, you know, the polarizing the two you get these nice phasic glycinergic currents, really amazing work. Nice to see you on the UK after a long time. Yes. And you actually answered the question that I was too stupid to think of. So my, my question was about these very direct synaptic contacts that you have in the complexes where you might, for example, have an amocrine cell that synapses onto the terminal, the relief very close to the release site, the bipolar cell that's driving a ganglion cell. So what you have there is the opportunity to sort of microminiturize very fast and simple processing. I say simple because there's not going to be very many other inputs coming in down the little dendrites that are feeding them. And it would be really nice to know how effective those complexes are at processing information. But of course it's an incredible technical challenge. I realized that. Yeah, yeah. You know, I, you know, Simon now. I've been starting to try to learn a little bit of computational neuroscience and how to use neuron and Python, and trying to get people in my lab to also get enthusiastic about that. Because we know the morphology of many of these cells, we know a lot of the ion channels that are there. So maybe trying to get really good realistic.