 Okay, I think the traffic is delaying some people, and maybe even our breakfast is delayed, but should be here at some point during the talk. So I'm very pleased that Dr. Zambath or Sam is finally here. We actually needed to delay this several times because Sam really wanted to come in person and see more and see people here in person. So Sam got his bachelor's degree from Berkeley, and then he moved to UCLA, and there he got his master's and PhD degrees 1996 and 1999, and these were in physiology. And then he did his, he did two postdocs, one in Stanford and one in University of Washington. And then before obtaining his first faculty position at University of Southern California 2005, and now he's a full professor in ophthalmology and neurobiology at UCLA. So he has directed neuroscience graduate program, and now currently he is associate director at Stein Ein Institute and chief of vision science division at the department of ophthalmology at UCLA. So Sam has contributed significantly to our understanding of how rod and cone photoreceptors both convert light into electrical signal, and then how they relay this signal forward to the retinal bipolar cells. And his more recent work is focused on elucidating how retina adapts to retinal degeneration. So how, what happens to the retina when photoreceptors are dying, and on the other hand also to the evolution of photoreceptor light signaling. And this talk this morning will be focusing on the retinal degeneration part, and then there will be talk at noon, which will be focusing on evolution of the visual system by comparing light signaling in lamprey and mouse retinas. So I have known Sam since I was a young grad student. I came to the meetings at Arvo from Finland, and I met Sam, and he has been always supporting my career, and we have had very lively discussions about science and other things. And from that I have learned that he really supports young scientists. And I think that's very important because that's how you really make the impact in long term. And thus, it's not a surprise that he's also directing NIH funded vision science training program at UCLA. So I'm not taking more of your time with that, Sam. Please, the stage is yours. It's really a great pleasure for me to be here today. As Franz had mentioned, this was planned a couple of times in the past in the last calendar year. But I was really keen to come see Moran and meet with everybody. So despite all this white stuff on the ground, it's still a great pleasure to be here. So what I'd like to do today actually is tell you a little bit about one of the research areas that we've been working on in the laboratory over the last, I want to say three or four years. Traditionally, as Franz indicated, my interest of all started off in photoreceptors, understanding the mechanism of phototransduction in both retinal rods and cones. And then it sort of extended into understanding how these signals get propagated within retinal circuits. But, you know, over the last, I would say, you know, three to five years, there's really interesting implications that come out of basic work on photoreceptors. And namely, we can begin to evaluate how photoreceptors began to change during processes of retinal degeneration. Right. And this is what I'd like to share with you today. I'd like to share. Thanks, I'd like to share with you. Perfect. I'd like to share with you some of the work we've been doing particularly on retinal cones. So, let me, this is a figure up, you know, I'm sure we all know well here. This is the functional organization of the eye. And of course what the eye is doing is it's a sensory organ, it's taking light that reflects off of objects in the environment, and it's projecting it. I'm not going to have a laser pointer on my computer is a little bit glitchy. But if you can see the cursor here, it takes these focusing elements and focuses on the back half of the eye, which is where the retina is. And the retina is a multi cellular tissue, where the back most layer that a photoreceptor cells is responsible for initiating our visual experience. And the receptor cells are obviously very important for vision. They are the sole mediators of most of our conscious visual experience and so all of the information that's present that allows us to see with the complexity that we do has to be encoded within the photoreceptors themselves. And so understanding how photoreceptors encode all the salient features of our visual experience is actually quite critical in eventually figuring out how it is that we can have this robust sense of seeing. And as this robust sense of seeing starts to decline for instance during age or during retinal degeneration, we can begin to evaluate how the photoreceptors are changing. Now, I'm going to keep going to the wrong computer. One of the things that happens as many of you know, you know, being here within an I Institute is that photoreceptors because of the sensitive nature of their structure, namely this tightly regulated outer segment that contains all of the elements that perform phototransduction that perturbations to this environment can lead to photoreceptor debt. And this has been well documented, including here by seminal work by Robert Mark and Brian Jones. So when photoreceptors die, the vast majority of the reasons why, particularly for inherited retinal degenerations are due to the death of rod photoreceptors first. So as you can see right here, the typical trajectory of many retinal, of many retinal degenerations involving photoreceptors starts with the rods. And as they began to die, it reaches a threshold. And at this threshold, this is where cones began to die. So during this cell death, or the deaffrontation of the retina, there are several secondary changes that that occur. And I said they've been well documented, even here at Moran over the years. And these include in response to death, the neurons in the glia react with a stress response effective. So this is an additional gliosis. As the retinal mule cells try to encapsulate the size of injury, the site of injury. And what happens beyond that is that the retinal circuits that carry photoreceptor input began to look for the glutamatergic input that's lost from the rods. So bipolar cells will send aberrant processes into the outer nuclear layer. Inter-retinal circuits will rewire to a modest extent, but they still rewire. So what's been suggested through the years is that these remodeling efforts might hinder the ability of the visual system to be recovered, following various forms of retinal degeneration. So this is something that I've been really interested in, and particularly in relation to the cones, because as I mentioned before the cones are the cells that actually mediate most of our visual experience during the day. So the cones actually undergo these well-known stereotypical changes and structure as well. What happens is initially, this is a cone, I'm just going to share with you the different pieces of the structure here. This is the light dependent outer segment, which is created by the continuous imagination of the plasma membrane. This is where phototransduction is housed and packed in. It's the highly efficient structure of the outer segment that allows phototransduction to be so efficient, both in rods and cones. On the other side, there is a synaptic pedicle, and this is the site of contact between the photoreceptors and all of the bipolar cells and horizontal cells. But what happens during the stress of photoreceptor degeneration is that these processes begin to retract. They initially get pulled back into the cell body. And what you're left with, when a lot of rod photoreceptors begin to die, is you're left with basically a cell body and some of the inner segment of the photoreceptor. And these have traditionally been called dormant cones. Dormant, because I think the assumption has always been that how can a cone that looks like this be functional. And so this has been the target for many therapeutics, because interestingly enough, and even in many human retinal degenerations, these dormant cones can exist for years before they themselves die. And so the question is, what are these dormant cones? What are the changes in their membrane properties that are contributing to this phenomenology? And what can we learn about this process? One of the interesting things that's been found about these dormant cones, and this is data that we've known for some time, is that when people record from them, their resting membrane potential is fairly depolarized. It sits actually pretty close to the resting membrane potential of normal photoreceptors. And that's very surprising, because most neurons in our brains sit near what's called the potassium leak potential, which is around minus 60 or minus 70 millivolts. And what happens is that the outer segment and the cyclic nucleotide gated channels in the outer segment are responsible for dragging the membrane potential into the depolarized state. So if there's no outer segment here, it's curious about what the mechanism is that causes the depolarization. And this is one of the things that we began to think about. It's like, how do the membrane properties of these degenerating cones change? And can this give us some clues about the mechanism? Well, interestingly enough, there is a mouse model that we can use to study this very problem. The Jackson Laboratory contributed many different retinal degeneration lines to the literature. Of them, I think the more famous brother or sister of this model is the RD1 mouse. And the RD1 mouse is a very aggressive retinal degeneration. To be honest, I think it has relatively limited utility because the degeneration is so rapid that it's included also by the normal development. And so it's hard to disambiguate development from degeneration. This is where the RD10 mouse is actually quite nice. This is already retinal degeneration line 10. And what's been found in this mouse is that the degeneration is much more gradual. What the RD10 gene is, is if we took this piece of the rod outer segment right here, and looked at it more carefully, what the RD10 is, is a mutation in the PDE6 gene. So this is the cyclic GMP phosphodiesterase, which is the effector molecule for phototransduction. And what happens here is that when it's eliminated or its expression is decreased, you end up with this very slow degeneration that's been characterized previously. Under normal conditions, what you see is this is, of course, you know, a cross-section of retina. We have a stain for peanut aglutinin, that's the PNA there, that stains the sheath of the cone extracellular space. And you can see cones here, and you can see all these rod cell bodies. This is the pithidium, but what happens over time, over eight or nine weeks, is that the rod photoreceptors begin to die. And what you're left with at this time is the single row right here of dormant cones. So it provides an interesting way of considering what these dormant cones are doing, because we can record from them physiologically. Now one of our original, oops, one of our original hypotheses about what's going on with these cones is that it could be that during the stress of degeneration that there's some type of metabolic insult. And the metabolic insult basically has deleterious effects on the cone structure. And we have some insight about this largely because we've made calculations of the energy expenditure from a cone. So by characterizing the membrane properties, we've been able to calculate how much ATP is consumed by the cone as a function of background light. And the reason that we can do this is because in neural tissues, the vast majority of the energy consumption actually occurs because you need to pump ions against their concentration gradients. And in fact, that's why the brain is so metabolically active as well, is that all of this pumping activity creates this need for ATP. So we've calculated the amount of ATP required for cone photoreceptors as a function of background light. And this is from a paper from a couple of years ago. But what you can see here is if you took the total amount of ATP here, you can see that the amount of ATP consumption actually goes down with background light. And this might seem counterintuitive, but this makes sense because what phototransduction is doing is it's closing cyclic nucleotide-gated channels. And by closing channels, you're reducing the amount of influx of ions into the cell, which means the need to pump it out decreases. As it turns out, when you consider the membrane properties of a cone, there are two conductances that dominate the energy expenditure. The first is the cyclic nucleotide-gated channels, which are CNG here. But the second, interestingly, is the synaptic calcium channels that permit the release of glutamate. Now, in rod photoreceptors, this is flipped backwards. Rod photoreceptors, this ferial typically only has one ribbon and one side of release in the malien retinas. But in the cones, there can be as many as 40 ribbons present in the periphery of something like, you know, a cat or rabbit retina has been shown by Helga Kolb here and others through years. Right, so when you look at this, what comes out of this is that these two conductances dominate the total energy expenditure. So if for some reason there was some metabolic insult to the retina, it might make sense that the photoreceptors would try to retract an outer segment in a synaptic pedicle with the presumption that they could be trying to save energy because without an outer segment in a synaptic terminal there, there may not be the same need for those ions to be pumped out. So we looked carefully at this in the context of several conductances here. And the experiments I'm going to show you that involve the physiology of recordings from the retina directly were done by an I-star resident, Erika Ellis, who's recently finished her PhD in the laboratory and gone off to her residency right now. She's a PGY2 right now. But what she did is she did these really heroic experiments on the RD10 retina, where she would take the retina from a dark adapted mouse at eight or nine weeks, where all the rods pretty much have degenerated. She would take the retina out. She would embed it in an auger. She would cut slices. You can imagine how difficult this would be under conditions where the retina is kind of crumbling apart. And then what she would do is she would make wholesale voltage clamp recordings from these cones. And again, the virtue here is that there is a single row of cones. And based on this, she was able to measure and calculate a number of things. This includes measurements of the passive membrane properties that include the capacitance, which is a measure of the total amount of membrane. Also looking at the resting membrane potential as well. But what she was also able to able to evaluate based on methods that we've developed in the lab over the last five years, is looking at all the other major conductances in the cone as well to see if they're present and to what extent they are participating in the membrane properties. And these include the outer segment cyclic nucleotide or CNG channels. It includes on the other side, the calcium channels at the synapse. But it also includes other channels that are involved in stabilizing the resting membrane potential to come. There's two in particular that we've been curious about. One are these HCN channels. This represents a hyperpolarization activated non-selective cationic conductance. And on the other hand, it's a sustained voltage-gated potassium channel that's typically called IKX for those of you who follow this literature. And what happens is that the HCN1 channel and IKX work together to sort of hold the resting membrane potential like really close to the resting membrane value of like minus 40-ish normals. So we became curious about how all these things were changing in dormant columns. So this is what one of these recordings looks like on the right. So what Erika has been able to do is she has been able to fill the cells during recording with a die. I called Alexa 750. So you could then image it in the red. If you don't get a fantastic image, you can see all the key features of the tone here. And that includes the outer segment, which is this tip out here, a synaptic pedicle on the other side, and of course the cell body. But what you see is during degeneration, even at the earlier times when there are still raw photoreceptors left behind, this is only at four weeks, you can see that it's become this dormant cone form. And in fact, if you were to wait even out to eight or nine weeks, what she began to see in these cones is that they would start to show these external perturbations that we, I'll describe more about what these might be doing. But what comes out of this is when you measure the capacitance is that the capacitance, even at the earliest time points that we could look at was already fallen to a value that reflected the loss of the pedicle in the outer segment. And that what she observed is over time, that there was a gradual increase, presumably because of these external perturbations. Okay. So, when she looked carefully at the resting memory, she found what was consistent with what other people had found in these cones, that there's no change in the resting memory and the function of H in degeneration. It looks like they are all, you know, roughly in the same range of sort of minus 15 millivolts roughly. Right. And so, so this is actually quite interesting. If you don't have an outer segment, and you don't have a snap to pedicle, what is it that's keeping it at this reasonably depolarized potential. We did a preliminary test where I blocked the two conductances that I said that are typically responsible for clamping and resting memory potential around the value of minus 40. When you use either external or external cesium to block these conductances, you don't see an effect on memory potential. So, the question is, is that what what could be causing this. One possibility is that there could still be residual cyclic nucleotide data channels. What Erica looked for is she looked to see whether she could see them. Indeed, what one of the things that we found that was fairly that that's never been observed before, is that when you record from these cones, you can actually mention light evoked responses. You can see some variety of what the light evoked responses look like on many from many cells right here. This is just the cell that's that's imaged on the right is shown here in the dark. You can see that even with the brightest light we can deliver we can't saturate this current, so we don't know how big this current actually is. This is in contrast to the wild touch condition again where you can see that the size of the response to the cone is quite large. In fact, in this case, you can see a cone which has a cone response followed by the rod response because of gap junctional coupling between rods and cones and the retina on the dark adapted state. So the fact that you can see a light response here was quite interesting and it sort of raises the question is like what's causing the light response and can we see the cyclic nucleotide gated channel current. Well, one of the things that could be causing the light evoked response is that it could be the normal photo transaction mechanism. It doesn't seem to be developing into the the psyllium doesn't appear to be generating a normal outer segment, but that doesn't mean that the cell isn't trying to put these things into the membrane. And in fact, our collaborative work with David Williams at my institution, along with his postdoc Antonio and our EM Tech with Punjau looked at this more carefully to a combination of super resolution microscopy and also by EM. And what we find in these cases are a couple of different kinds of tones, depending on whether you look centrally. In many of these tones centrally, you actually see that there's a cell body there, and that there's no perturbation of the of the psyllium to create another segment. But when you stay with an ops nanobody, you can see that the ops nanobody is now all over the cell box. And an example of this tone EM can be seen, not this particular tone, but a similar tone can be seen right there. But what we see as we move toward the periphery is that the psyllium seems to be generating some amount of membrane. It's not a formal outer segment, as it were, but it is creating a structure that elongates, and you can see that there's ops in there as well. And you can see an example of this shown by EM like this. Right, so there's ops in there. So there's obviously the substrate to absorb light, and to be able to then generate a light response. We did some preliminary experiments to try to block those channels with a drug called elsis diltiasm. As it turns out, that was probably a bad idea. Elsis diltiasm, while it's an effective blocker of the rod cyclic nucleotide gated channel, doesn't do some well with it for cones. So we opted for the opposite approach. So instead of trying to block the channels, we thought, okay listen, since we have access to the inside of the cell, what Erica could do is she could then add cyclic GMP inside the cell and see if she could have all the cyclic nucleotide gated channels, which is what she did right there. This is an example of one of these recordings. So what happens at time zero is she breaks into the cell, allowing the influx here of, in this case, 100 micro molar cyclic GMP. This is a supersaturating concentration of cyclic GMP. You should open all the cyclic nucleotide gated channels, and you can see the slow diffusion into the outer segment creates this current that opens up over time. Right. So just whether the same could be done for the dormant cones. And so we did, we did this, she did the following experiment. Here, when she delivered a series of flashes first, this was a cone that had one of the bigger responses that she reported from this case it was about 10 people. But what she observed is that as she broke in, a current still developed. And the current, you know, was, you know, probably a half or a little bit more than half of the current that could be generated from the normal flow. And in fact, you can get to a level here where you can saturate the dormant cell body with so much cyclic GMP that you don't get light responses anymore, presumably because you're saturating the channels. And whatever residual phototransduction mechanism there is there, can't no longer create a change in the number of open channels. So, just just for a little bit of mouth here so we have about half the size of the maximum cyclic GMP gated current that's possible. We also have about half of the total membrane. And so when you put these two things together, what it suggests is that the density of cyclic nucleotide gated channels and the membrane of these dormant cones may not actually be that much different from what you see in the normal cone under regular conditions. Right. And so, I think that the conclusion from this work is that, look, it looks like there's cyclic nucleotide data channels there and those cyclic nucleotide data channels are supporting a conductance that can keep the resting membrane. What happens, okay, with depolarization, what is it due to the other side, right, there's no pedacle there but is there a calcium conductance. So Erica measured this by under conditions where she can isolate the calcium current, and then she delivered a voltage ramp, and so that she could look at the current voltage relationship of this current and an example of that is shown up here. And these are, you know, errors across both. What you see here is that, interestingly enough, with no pedacle, there seems to be a calcium current that has the same magnitude and voltage sensitivity as the normal cone. Right. And so you can see this right here. So as you ramp you begin to open calcium channels and then the calcium channels that the driving force for the calcium current begins to decline. That's the nature of this curve right here. But if you were to look at the distribution of cones that she recorded from both in the RD10 and blue and wild type in black, you can see that they essentially overlap. If you have a normal size calcium current, and you have cyclic nucleotide data channels there, I think this speaks against the hypothesis that the retraction of membranes, the result of some type of metabolic stress. It might actually be much more complicated, because based on these values, you would still expect the energy dependence of this cone to be quite high. Just for completeness, we looked also at the hyperpolarization activated conductance, and she can actually see no difference between these two either. So it's not that the compliment of channels between a normal cone and one of these dormant cones is really that much different. And we found that to be quite surprising a result. What was even more surprising is when Erica looked downstream at bipolar cells. When she records from the bipolar cells, she can see all of the responses on the left from an on bipolar cell. This is her fill. You can see that the on bipolar cell is extending an axon down into the proximal half of the interplexiform layer. You can see it from an off bipolar cell, which sends axons more shallowly into the interplexiform layer. You can see it from a horizontal cell, which you can see extending processes laterally. In each one of these cases, we know that no rods are present, because you can't see them here. You can sometimes see patches of retinal pigment epithelium right up against the cone. I should note though, and I didn't say this before, this is an important point. The light responses that you see in the cone appear to require somewhere between 100,000 times more light. And that's also consistent with what she sees in the bipolar cells. So while you can generate response families here, these are again voltage clamp recordings where she's plotting the current as a function of time and delivering a flash at time zero. The sizes of the responses that she sees in the bipolar cells are actually not too dissimilar from normal, except for the fact that they require about 100 to 1000 times more light. And this was interesting, because the cones themselves must be transmitting this information. What this means is, is that the resting membrane potential must be driving a rate of glutamate release that can be suppressed in light to create the profiles that we see for both on and off bipolar cells. So this is actually quite amazing, because what it indicates is that these cones are not dormant at all. And in fact, perhaps what the main strategy needs to be for patients that are suffering from rodent generation is trying to figure out how to protect these cones. Perhaps an enhancement of light sensitivity might be helpful, but protecting them seems to be the more important thing that we can do. So, you know, we were curious about what this did to the retinal output, the retinal ganglian cells. And so we collaborated with my colleague, Greg Field, who's about to move to UCLA from Duke, by the way. So, but he is known for his work using a high density multi electrode array. And you can see an example of what the array here. And so what he does is he takes the whole retina and he lays it flat ganglian cell side down on the array. And the virtues of the mouse retinas, it's only a single layer of retinal ganglian cells. And so what he can do with stimulation is he's able effectively to measure the properties of retinal ganglian cells by triangulating the action potentials that are generated and measured by each one of these electrodes. So the upshot of this is that he looked under various conditions in the same eight week old RD10 retinas. And what he found is that when you look under the Stotopic light intensities, what he sees, and this is now a distribution of all the retinal ganglian cells and the amount of information they're transmitting. And what he sees is that Scotopic light intensities that all of the RD10 retinal ganglian cells are displaying much lower information than what you would see under normal conditions. And this makes sense. It's Scotopic light intensities. You're driving rods. If you don't have rods, there's a whole lot of information, right? Going to the middle, it's very much the same. Scotopic light intensities, you have rods and cones. But remember, because the mouse retina is still 97% of rods and 3% cones, it's still dominated by the rod input to a great degree. And you can see that there is a loss here of information for the RD10 retinal ganglian cells compared to the normal retinal ganglian cells. And interestingly, what he finds is that there is no difference in the distribution of information rates at Scotopic light intensities, light intensities driven by cones between normal and RD10 retmets. And this is quite interesting. You know, it means that the retinal output to some extent is preserved. You can look at this even further by subdividing these things into their spatial and temporal components. And what he discovered is that, yeah, sure, there is a very modest shift in the area of the receptive field, which you can characterize on a cell-by-cell basis. You know, mounting to, you know, 500 microns squared among 8000. This is like a, you know, like a less than 10% change in the receptive field area. There is no change in the temporal receptive field. How he characterizes this is this, these panels here represent something called the spike triggered average. So for every spike in a retinal ganglian cell, he looks at the light intensity preceding that spike and averages that. And he can use the time it crosses the zero mark as a measure of the temporal receptive field and they look fairly normal too. The major difference between the two is when you look at the firing weight. So the gain of responses for the amplification responses in the case of these retinal ganglian cells is only shifted by about a factor of two. So what's fascinating about this is that you have outer retinal circuits that are desensitized by a factor of 100 and perhaps more than 100. But when these are now represented in the level of the retinal ganglian cells, the relative shift is smaller. And what we're thinking about this, this sort of masking of the sensitivity difference is that when you get to the inner retina, a big part of what the inner retina is trying to do is it's trying to do some type of contrast normalization. And so what it's looking for is relative changes and not absolute sensitivity, the way outer retina circuits are. And it could be that the normal mechanisms of contrast adaptation are in a way compensating for the absolute loss of sensitivity in the outer part of the retina. What's fascinating about this phenomenology is it appears to hold across multiple models of retina degeneration. So this is another example of a paper that we've done with Greg's lab, using a different model of retinal degeneration that is an elimination of the cyclic nucleotide gated channel beta seven. This degeneration is super slow. Okay, it takes over six to seven months for the rods to completely degenerate, but you can see the same phenomenology occurring in the cones right so on the left here it's the percentage of from wild tide and you can see that during degeneration the rods decline really quickly, the cones decline more slowly, but the cones are undergoing the structural changes. And what the upshot of this is, and I'm not going to go into a lot of detail into this, but when you look at the amount of information present in the retinal ganglion cells, in particular these on brisk sustained cells you see that the amount of information actually doesn't decline much until the cones really began to die. So whatever this might seem like, is that there may be common mechanisms, regardless of how quickly the rods are dying that allow the retinal ganglion cells to protect as much visual information as possible. I mean, in fact, this is a property of almost all neural circuits that homeostatic compensation has a great way that what it's trying to achieve is it's trying to achieve a sort of protection of the amount of information that's transmitted. And so we aren't sort of in the process of trying to look at this a little bit further and understanding, you know, what are the mechanisms that might protect sensitivity, especially as you get into the inner retina. So let me just give you some broad conclusions here. So the cones and the degenerate retina, you know, these dormant cones, you know, appear to display a reasonably normal dark resting memory potential despite the loss of the synaptic and the outer segment structures. There are a lot of light responses there, but they're smaller and less sensitive. They appear to express phototransduction machinery, right, even in a more abnormal cellular elaboration or through the inner segment. The second order cells look like they're light sensitive as well. They still display robust responses, although they need more light intensity, reflecting the lower response gain of the cones. So retinal ganglion cells at photopic light intensities, the spatial and temporal receptive fields appear to be marginally changed from normal, right. There's been a limited amount of degeneration of sort of remodeling, if you will, perhaps. And what this, and while the gain is lower, which reflects the loss of sensitivity of the cones, it's not as low as you might predict, right. The main change in light evoke signals in this is the reduced sensitivity with almost all the other parameters, you know, staying somewhat normal. And this is, I think, really encouraging. What it suggests is that the protection, as I was mentioning before, these dormant cones could be quite important for maintaining visual sensitivity. And indeed, you know, efforts like those from Botan Roska's lab to introduce optogenetic reagents into these cones to enhance sensitivity might be helpful. But I think the majority of this really is making sure that we can keep these cones alive. Now, before I say my thank yous and acknowledgment, I just wanted to give one final thought. So we have this degenerating cone structure here going from a normal cone to this dormant cone. But, you know, there's a lot of parallelisms if you look in the literature about what these dormant cones are, right. If you're to look at retinal organoids, for instance, retinal organoids, cone cells and retinal organoids display many of these same characteristics. They elaborate membrane from the psyllium, although they don't generate what we consider to be bonafide outer segments. They do elaborate membrane. They express all the phototransduction stuff. Recent work from Ronak Sinha and David Gann has shown that, you know, they also can generate light evoked voltage changes, right. In addition, you know, there's a phenomenology that we see in ground squirrels and other species as well. And this is interesting because, you know, during hibernation, you know, the cones undergo much of the same types of changes. They retract their synapse, they retract outer segments. And so kind of the thing that I've been sort of wondering myself and I'll just wonder out loud now. I haven't said this out loud before, but I wonder whether there are just common mechanisms that we need to consider. That could be helpful. You know, the ground squirrel is particularly interesting because from season to season, they will regenerate brand new cones from these dormant cones. And so the real issue is that, you know, are the dormant cones that are left in the retinal degeneration patients, is there some insight that we can gain from them that can promote a more normal structure under therapeutic conditions. And with that, I just, let me just say a couple of thank yous here. A lot of the work that I've done is in collaboration with a long time colleague, my PhD advisor, Gordon Fain, who I jokingly says my oldest postdoc. We also have within the institute, my colleagues, David Williams and his postdoc, Antonio, Greg Field, who performed a lot of the multi-electrode work. And really all of the patch clamp reportings that I showed you were done by Erica Ellis, who is a PhD student in the lab and is now a resident at the Stein Eye Institute. I'd like to, you know, thank you all and I can take questions, but I'd like to go in this order. I'd like to start with questions for anybody in the audience and then I can take questions from Zoom. Great. Great, yes. Just this animal, how do you know that you weren't recording? How do you know we're not recording from those? Yeah, they very well can be recording from those. Those contributing to the rental families. So I think what we see is one of the things that happens is that by virtue of recording from all of the rental ganglian cells, we can identify them all individually. So this is what I want to say. And so he based on his recordings from a patch of rental ganglian cells can see every single subtype. And so the IPRTCs are only a small minority of the total cells. And so they can't account for this. So we know that the IPRTCs are surrounding on it as well, but they can't account for what is here. And the IPRTCs are also much less sensitive to. So it would be perhaps of the topic range, but they can't account for what all the ganglian cells are surrounding. As there are only a couple percent total. I'm just curious, how much of this you expect to be specific to this particular genetic model of the generation versus how much it's done. So I think that was the last slide that I was showing. I actually think this is incredibly generalizable. So we just published this paper, right? So this just came out earlier this year. And this was that model that I was describing with my own generations very slow. So this is compared to a faster generation. The slower gene generation shows the same phenomenology in the cones. It's that despite the fact that you put that on the map, it still responds pretty robust. When you look at the receptive fields of the retinal ganglian cells at all of them, there is a decline in the amount of information when they begin to die. But it seems like there are mechanisms that are protecting the system in retina. We haven't looked at the outer retinal and bipolar cells in this model much. Only the rod bipolar cells, which when the rods began to die, the snaps began to pull apart, right? And so you start to lose rod to rod bipolar cells. So that's one of the features of a mammalian retina is that rods project when it's class bipolar cell called the rod bipolar cell. But the work that we've done in this model, this deficiency actually has a way to reverse itself, right? If you were to apply tamoxifen to this mouse, if it's made with the right preline, the stop in the cyclic nucleotide subunit comes out and you get normal expression from the normal locus. And so what we found is that those synapses come back. And if the rescue is fairly early within one or two months of that rod to rod bipolar synapses, this is collaborative work that Greg and G-Chan have done. So, but I don't know, but I think just getting down to the level of the retinal ganglian cells, it looks like the phenomenology is very simple. Despite the fact that there's a fast degeneration and a slow degeneration, which suggests that, you know, common homeostatic mechanism is probably the way. Second France. Yes. So this is fascinating. Love the lecture. And so part of what you've helped with is the mystery of why when we know the genetic defect is with rods, cones eventually die off as well. But we know another important factor with these degenerations is the complex rewiring that occurs over time that Robert Mark here, Brian Jones have shown. Do you have some information to show that as long as you have dormant cones in place that that apparent rewiring does not occur and only starts occurring once the cones died off? Or is there still rewiring occurring? That's an excellent question. So let me just add here is one of the things that one of the things that Greg saw in the retinal ganglian cell report is that with the so here's what we believe. Okay, because there's responses in the bipolar cells. I think what's happening at the level of the donor cones is that you're trying to stay depolarized so that they can maintain glutamate release onto the second order cells. And as long as the second order cells are seeing glutamate release, I think it triggers against the rewiring. And that's why when you look at the retinal ganglian cells, they show temporal receptive field. So the combination of the space over which they can be stimulated by cone photoreceptors, which is the factor on the left. But you only see a marginal reduction in that size. But importantly, I think that the contraception tells you a lot. What it means is that the wiring of these signals as it comes into the inner retina is not being modified in a way that impedes the temporal sensitivity of the retinal ganglian cell responses. So I suspect that's that's the right intuition is that real rewiring and re-bottling that you see might happen only at the stage where the cones are beginning to die. But as long as those dormant, I keep saying dormant cones, I don't want to say that, but as long as those cones remain, it may be protective for the rest of the surgery because they appear to be releasing glutamate that contributes to light responses. So for rescue, clinically, just to show you the point, it will seem to be it's critical to try to maintain dormant cones or rudimentary cones. I don't want to call them as long as we have those in place, rescue is much more likely if we have a wave. Absolutely. Once the cones are gone, that's when rewiring occurs. And that's where it gets to be way more complex and less likely to get great vision. And I think that I think that's excellent intuition. Yes, Francia. So in that model, you can reactivate the cones. So you have a dormant cone. We haven't looked. So because you said that they may be hibernating. So in that model, if you get dormant cones and then when you reactivate the rods, then the dormant cones going back to the normal cones. And the answer is no. So actually, we have a manuscript that we're working on right now, where we evaluated the recovery as a function of time across all these time points that you see here. You get to the point fairly late in the generation, like five months ish, where the rescue doesn't help anymore. It just keeps degenerating. So, you know, I obviously, I think it's much more complicated than this, right, but there is a point of no return. We're even trying to intervene. I mean, it's the reason why screening is so important, right? The brain and the brain extension of the brain is performable. It covers up everything until you have a problem. And by the time you realize this problem, sometimes it's too late. I mean, we see this problem a lot. It's the loss of the retinal ganglite cells is the, you know, you can sustain a pretty massive loss of retinal ganglite cells before you start to notice. Yes, Deepa? So amazing. We see very reduced photo response for the dormant cones, but we do equally well calcium and calcium current. So how does that translate to? What do you think? So because the calcium current density, as a function of voltage, looks fairly normal. I think this is a situation where even though the cones are dormant, maintenance of the resting memory potential, perhaps the somewhat normal release. Although I mechanistically don't understand how, because there's no pedagol there, right? You know, the thing that those axonal protrudes that I'm showing you early on, you know, there's some other work anatomically that seems to show, that they show a lot of markers for synaptic release, like steam to synaptagment and other sensitive markers. So I think this is a case where the sensitivity of the cell is less by virtue of the fact that there's not an organized average segment. From the perspective of the rest of the circuitry, it seems to be receiving glutamate in a way that may not be too normal. Related to this is kind of easy for me to believe that this is normally called still express CNG channels and it's kind of like IPRGZs. It's just don't have the outer segments of those kind of audience. But then, it's very hard for me to believe that this synaptic transmission is so normal, because you don't have the pedagol. And then, like, this is kind of like in neuroscience, you have the synapse where you need to have like very, very small distance between the pre-synaptic side and the post-synaptic side. So how business is going? It must be that they, look, the fact of the matter is, is when you look at the temporal profile of the response, the temporal profile doesn't seem to be normal either, right? You're getting responses that activate in terms of matter of, you know, hundreds of milliseconds, right? So whatever is left over, still able to create, I think, what are probably fairly normal synapses, the structure may not be the same. But they appear to be, like I said, in order to get on and off responses on the right, there has to be sustained glutamate release, and that sustained glutamate release has to be suppressed by the right. I think that's the only explanation that makes any sense. I don't know if Brian is here, but I think EM, like doing EM, what that's like. I think that we're in process, so David Williams is, we're continuing work on those, but no, I think that's from you. I think understanding how those synapses are regulated. Are there any questions on Zoom? Okay, how is that clear? All right, well. I have one. Yes. So I'm not sure if I understand where you kind of, so you have this metabolic hypothesis that is derived by ability factor, which is kind of supposed to promote glucose entrance into the cones. And now I kind of understood that you maybe didn't believe that. No, I believe that. I believe that, but again, I think the point, I think the simple hypothesis that it's metabolic is not, that's an explanation. I think it could partially be metabolic as well, but I think it's more complicated than that. So, so I think what Francis referring to is there's this beautiful work area in Paris that shows that rods actually secrete a trophic factor called RDSCVS rod derived cone viability factor. And it interacts with a receptor on the cone membrane that promotes glutamate uptake. So it might be an important part of the maintenance. The reason that there's a second, it proposed this is the reason why there's secondary cones is that the loss of rods leads to the loss of trophic factor that keeps the cones alive, because that's not there to promote. Wow, that's pretty cool. And then the point was to use for the full well cones, because they are not surrounded by the road forward. So I may somehow be friendly. And maybe that's why they are not sensitive in any disease. That's possible too much, but I mean, but that the meal or glia, you know, could be, you know, or another source of metabolism. We know that the cones use the glia after their visual cycle. It's not maybe crazy to believe that, you know, particularly the foveal cones are using a different source for, for, for the septaic metabolism. And they're the ones that persist the longest, that's right. So they obviously probably have, they're different enough at the loss of the rods and not as big of impact. Eventually they'll go to, but they will often that little tiny center vision will stay and stay a long time. Yeah, that's right. That's what makes these so difficult, is that I think people still notice, right, until, until their tunnel, right. Fantastic. Thank you.