 Είναι ουσιακό να μην κατασταθείτε ότι είμαστε αυτοί και πρόκειται να δούμε έναν άλλο πρόσιο της ΑΕΣΕΚΕΙΣΕΡΙΣΕΙΣΕΕΙΣΕΕΙΕΣ. Η πρώτη από αυτή η 4η σύζυνη και όλοι όσο στη δημιουργία του κόσμου. Είμαι George Café-G, ένας δημιουργείο της Σόλας Τόμα και τώρα είμαι ένας παιδιά της Σόλας Τόμπα. Στις your hosts for today, I would like to once again begin by thanking Tim Vogels, Panos Bozellos, for putting forward this ever-expanding initiative towards a greener and much more accessible seminar world. Having said that, allow me to get back to the reason we all gathered here for today and introduce our guests from University of Wisconsin, in Madison, Dr. Ronak Sinchak. Ronak did his undergraduate studies in human physiology at the University of Calcutta, followed by a year in the Tata Institute of Fundamental Research. στο μέρος της επεισύτητας της ήρωσης, οπόρ expense was the International Max Planck Research School, where under the supervision of Jurgen Klingauf in the membrane biophysics department Roanak studied properties of synaptic transmission in hippocampal neurons. In 2012 he joined Fred Rieke as a post-doc at the University of Washington in Seattle, and there he remained for some very fruitful years as a senior fellow για να μιλήσουμε το 2018 στο Βουσκανάτος και να ξεκινήσουμε τη δυνατή του σχέδιο του. Μετά από το χρόνο με τον Φρεδ και στις τώρα, ρόνακ έχει been focusing on retinal processing at both the cellular and the circuit level and by studying forvated organisms, primates, he has identified novel mechanisms employed in central but not peripheral vision. So in his talk today entitled the regional variation of photoreceptor and circuit function in the primate retina, we will be hearing about the latest and I'm sure exciting findings and without any further ado from my side, please all welcome Dr. Sincha. ρόνακ, the stage is officially all yours. All right, thanks a lot George for that beautiful introduction. Okay, let me share my screen. All right, can you see it? Yes, all good to go. All right, all right. Thanks again George for that nice introduction. Thank you all and thanks to the Sussex Visions group, Dom and his lab for creating this wonderful virtual seminar series, which has been an incredible resource and I4-1 has been a great beneficiary of this virtual seminar series throughout the last two years. So today, it's really exciting for me to share a couple of unpublished stories with both of which we focus on regional variation of photoreceptors and circuit function in the non-human primate retina. So I typically show this slide to kind of invoke how the wide range of visual inputs over which our visual system operate. But today, I'm going to use this slide to talk about the two parts of my talk, which is the first part I'll focus on telling a little bit about functional properties, how they change across the visual field with respect to cone mediated signaling. And in the second part of the talk, I'll tell you a little bit about how our raw mediated signaling, especially the synaptic connectivity and organization of the very first synapse between the raw bipolar cell change as a function of space from the center of the visual field out to the periphery. All right, so as we all know that our visual system doesn't sample the visual space or the natural environment around us homogenously across the entire visual field. And this is illustrated over here as we are using our central high resolution or high equity vision to focus on the B on the sunflower, to the butterfly on the sunflower. And this is the region which has the highest spatial and chromatic resolution and which drops precipitously as you go away from the center of gaze towards the periphery of your visual field. Now, this is because we have this very unique specialization called the phobia, which among mammals is unique to only dinal primates. It is in this fundus image, it's shown in this white arrow, which is about less than 1% of the total retinal surface area, but accounts for about half of the retinal output. And this is on the top right. You can see that textbook picture of the rod and cone density map as you go away from the center of the phobia out to the periphery. So eccentricity is degree at visual angle in monkeys and humans, one degree of visual angle is roughly around 0.25 millimeters. So you can see that the cone photoreceptor density is the highest in the center of the phobia and that's also where rods are completely lacking. And you can see this cone photoreceptor density drops dramatically as you go out from the phobia, so zero degree centricity out to the periphery. And this is on the bottom as a top down view of that cone mosaic and you can appreciate the very densely packed cone array. So together with this high density of cone photoreceptor in the center of the phobia and a very specialized circuit called the midget pathway where you have a single cone photoreceptor providing input to a single on and off second order midget bipolar cell, which then in turn contact a single on and off midget ganglion cell. You have a private line of communication where the ganglion cell is sampling at the level of a single pixel, the pixel being a single cone photoreceptor. And as a result, you have this high spatial resolution. Now, as I said, this both spatial and chromatic resolution drops as you go away from the visual field. But if you think about temporal sensitivity or our sensitivity to rapidly changing visual input such as a flickering light, actually that goes in the opposite direction. So you have the lowest temporal sensitivity in the center of your visual field that's for your central vision. And that then progressively increases as you go away from the center of the visual field out to the periphery. And this was demonstrated first almost a century ago where the observer was put down and presented with a flickering light. And the frequency of that flicker was increased to a threshold at which the flickering light appeared steady and continuous. And this threshold value is defined as the critical flicker fusion frequency which is plotted here on the y-axis. And when you do this across the visual field, you can see this graded change in the critical flicker fusion frequency. And this is exactly the reason why the refresh rates of monitors today are set at 60 hertz so that we don't perceive that flickering light. And I find this quite fascinating because this has sort of informed the evolution of movie frame rates. We started off in the era of silent movies where we had a frame rate of 12 hertz, 12 frames per second which would appear like sped up and which were within the below the critical flicker fusion frequency. This was then when the introduction of TV moved to 24 frames per second, 60 hertz now with the refresh rates of monitors where we are trying to avoid this flicker perception and nowadays we can even go higher with the refresh rates to about 120 hertz where you get this absolute perception of fluid motion. Now this difference between peripheral and foveal vision in temporal sensitivity is almost twofold. And for the longest time we didn't know where did this originate in the visual system. And about a few years ago we figured out that this originated at the very front end of the visual system in the cone photoreceptors themselves. So if you measure the response of a cone photoreceptor in the fovea and compare that to this peripheral counterpart to a brief flash of light you know you can see that the voltage response shown here to a brief flash of light presented at time equals zero is about two times slower compared to that in the periphery. And this was fascinating because immediately we thought that you know this has something to do with the morphological specialization of the cone photoreceptors in the fovea and that morphological specialization was also known for a long long time and you can see this cross-section now of a cone photoreceptor in the fovea which is has long outer segments, inner segments and has these really long axons which can span over a half a millimeter in length. So we thought okay having this long axon probably serves as you know to temporarily filter out the signals. But turns out when we did you know more experiments you know using pharmacology, did voltage clamp experiments we found that this long axon is really not the reason for this two-fold slow kinetics of the foveal cones. In fact this difference in kinetics originates in the phototransduction cascade itself. So you know these experiments you know sort of invoked a bunch of fundamental questions about regional variation of cone mediated signaling in the fovea and in comparison to the rest of the primate retina. And there are four salient aspects that I will touch upon in the first part of the talk just like the temporal sensitivity that we I showed you which was measured using the critical flicker fusion frequency at the behavioral level. The first question I'm going to ask is whether this changes in temporal sensitivity or kinetics of cone photoreceptors is graded as you go across the visual field from the fovea out to the peripheral retina. Now we probed cone kinetics in these initial studies across a single background luminance and we wanted to see whether this change in kinetics persist across a wide range of background lighting conditions. And third we wanted to understand okay the kinetics are different between the foveal and peripheral cones. What about their magnitude of their signal? What about their cellular noise or intrinsic fluctuations in their responses and and how does that impact the absolute sensitivity or detection threshold of cones between foveal and peripheral retina. And we know that adaptation relies on both photoreceptor and circuit mechanisms. And I'm going to ask in the final question in the part one is what is the relative contribution of cone versus the circuit mechanisms especially in that specialized pathway the midget circuitry towards luminance adaptation in the fovea. All right so to do address all these questions so we're going to use good old patch clamp electrophysiology. Now a big shout out to the Primate Center the Wisconsin Brown National Primate Center and to their tissue distribution program which has allowed us to carry this research on fresh isolated retinas. Most of the data will be on macaque retina but I'll show you a little bit of the data on marmoset retina which also has been very interesting for us. So we isolate the retina put it on a dish and we patch from the inner segments of the photoreceptors. We present light from underneath using three different LEDs which different spectral sensitivity to match the short, medium and long wavelength options so that we can preferentially excite them. And here I'm showing you some exemplar voltage responses to a brief flash of light from a peripheral cone across five different trials. So the first question is do the cone kinetics change in a graded fashion or do the foveal cones and peripheral cones form to discrete populations. So to address that we look at an intermediate location which we call as a central retina which is around 1.5 to 2.5 mm from the center of the fovea. So we are going to present a brief flash of light at background light levels which saturate the rods and these are some exemplar responses at each of these three locations and you can already start seeing that the kinetics of the cone responses are graded as you go from the periphery which is in blue to the fovea which isn't red over here. And when you look at the average responses across the three different locations you can see that the foveal cones have the slowest response kinetics. The central cones have an intermediate response kinetics whereas the peripheral cones have the fastest response kinetics. So the cone kinetics do change in a graded fashion across visual space. So we want to validate this with another classical stimuli that has been very heavily used in behavioral studies the sine wave modulating stimuli where we change the frequency of the sine wave and a cone if it is slower will impact its frequency tuning curves and you can already see that the foveal cones at these higher frequencies of 32 hertz are you know do not track the signal very well and you can build these frequency tuning curves and you can see for the peripheral cones which have faster kinetics they're able to track these high frequencies and I'm plotting the gain on the y-axis and you can see the central cones their gain falls much in the middle between the foveal and the peripheral cones and so to quantify this we calculate the frequency at which the gain falls by an order of magnitude and that's the lowest for the foveal cones and the highest for the peripheral cones and for central cones that's right in the middle. So this is again validating what I showed in the previous instance that the cone kinetics do change in a graded fashion across the visual space. All right so we wanted to see if we can you know sort of replicate these findings in macaque retina in marmoset retina and one of the very interesting things in the marmoset retina is that the peripheral retina of the marmoset is cone dominant much more cone abundant than the macaque peripheral retina. So we again did the exact same experiments where we are comparing responses to brief flashes of light and you can see that this change graded change in kinetics is very much also present in the marmoset retina and you can analyze or estimate the time to reach peak amplitude of these voltage responses to brief flashes and you can see that very nicely is graded as you go from the fovea out to the periphery. All right so this was again at a single background luminance and we know that our visual system encodes over a extensive range of background luminance of about over 10 log units from dusk until dawn and we know that you know as light levels increase it has an impact on the temporal sensitivity and this has been very nicely demonstrated at the behavioral level again going back to critical flick of fusion frequency and I want you to focus just on the L and M cone mediated part the S cone mediated part is another story but what you can see over here is that when they plot this measure of temporal sensitivity as a function of background luminance it steadily increases with background luminance so and this is very well characterized also in vertebrate retina even at the level of cone photoreceptors their kinetics accelerate with background luminance so we want to see that whether this acceleration of kinetics is a similar or different across different locations in the macaque primate retina. So let's start with the peripheral cones where light adaptation has been looked at before here I'm again showing you voltage responses across different background luminance going from low to high and on the right I'm showing you a summary data where I'm plotting the average time to reach peak amplitude across cones at each of these three different locations and what you can easily appreciate is that the time to peak falls as the background light increases demonstrating that all cones irrespective of where they are located they've speed up their responses but the kinetic changes across eccentricity or across location is quite similar which leaves the differences in their time to peak or kinetics relative constant at a given background luminance so these cone kinetic differences do persist over a wide range of background luminance okay so kinetics what about the amplitude or the magnitude of the light responses how does that compare across different locations so we again look at the response amplitude we divide that by the strength of the light flash and so it's kind of a single photon response and you can see that across all these different background luminance the foveal and the peripheral cone amplitudes are their magnitude of responses fall right on top of each other there's practically no difference so if that be the case you have a scenario kind of like this right forget about the y-axis over here you have the same amplitude magnitude of the light responses which means that having a slower time course the area under the curve or the integrated responses of the foveal cones is going to be much larger right so if we look at the area under the curve and compare that across all the different background luminance for the majority of the background luminance you start seeing differences as expected right when you're calculating the integrated responses so the foveal and peripheral cones do have different integrated responses just because of the fact that foveal cones have a longer time course so everything that i'm showing you here is in some sense a measure of adaptation right so how do we formally describe this can we fit this with a certain function to kind of look at the decrease in gain as a function of background luminance and this has been very well described classically using the Weber law right i mean the effect of background light on behavioral sensitivity has been classically described by plotting the strength of the just detectable light flash against the intensity of the background light and these are these famous tbi curves or threshold versus intensity curves and where you can see that for the cone mediated visions i want to focus on this a substantial part of the curve has shows a proportionate relationship between the threshold of detection detecting that you know flash the dim flash at that background which increases as a function of background luminance now if you flip this curve you know take one over threshold which is sensitivity or gain you have an inverse relationship and you can also see then that sensitivity or gain decreases proportionately with background luminance now sensitivity has two things wrapped up in it it's determined by both the amplitude of the signal and the intrinsic noise or the fluctuations in current or voltages which over which the signal has to overcome to sort of adapt right so let's first look at the signal gain how does that change so i'm going to show you the same curves over here and fit them now with a weather function a modified weather function and this is the peak amplitude and you can see that they are very well fit with a weather function where you can see the gain decrease proportionately with background luminance and you can do this across individual cells and sort of calculate the background or the light intensity at which the gain falls by half and when you compare that for peak amplitude there's practically no difference between foveal and peripheral cones now let's do this for the area under the curve the integrated responses so if we just compare the peak amplitude luminance adaptation appears very similar between foveal and peripheral cones but when you do this for the area you start seeing differences so the peripheral cone integrated responses adapt much more strongly with background luminance than the foveal now if you do the same fit and calculate that half max value the foveal cones have a much higher value at which they adapt or the gain decreases by half compared to the peripheral cones so if you look at the area under the curve there are differences in luminance adaptation between foveal and peripheral cones all right so that was signal what about noise visual sensitivity is determined both by signal and noise and there is a history of measurements of signal and noise within photoreceptors which provide very interesting insight into perception and this is especially important for foveal vision due to their impressive sensitivity at a spatial and chromatic level and I'm showing a couple of examples over here for instance the ability to discriminate this rich palette of colors that we experience is dependent on discriminating changes in wavelength that is 50 times smaller than the width of the cone spectral sensitivity curves and if you think about spatial sensitivity that is about 10 times finer than the spacing between the cones so characterizing cellular noise in cone photoreceptors really is essential for understanding the limits of the cone noise puts on or imposes on behavioral sensitivity so how do we isolate cone cellular noise for this we are looking at photo currents we're doing voltage clamp recordings at different background luminance and the recorded current fluctuations include noise arising in the outer segment or the inner segment conductances and also include noise that are being generated by the instrument itself so instrumental noise now if you expose the cones to a near saturating light step as shown over here you close down most of the cyclic gmp gated ion channels in outer segment and you decrease the recorded current and you mockly decrease the current fluctuations so this way you can sort of isolate just the noise that is you can eliminate or measure just the instrumental noise and isolate the cellular noise so then we characterize noise by calculating average power spectra and isolate outer segment noise by subtracting the spectrum in saturating light from each of those background luminances all right so we record this cellular noise at across location and across a range of background light levels and we compare the measured noise to the dim flash response and you can see that frequency of the dim flash responses about 2 to 12 hertz roughly and across location the frequency range of cellular noise extends way beyond just the flash response range so we quantify rating over two frequency range one which we define as the flash response range which is from 2 to 12 hertz or 2 to 16 hertz and then a high frequency range and we refer to these as the flash response and the high frequency ranges now previous work on peripheral cones have measured cellular noise and in darkness and across a range of background light levels and on dim background such as this the noise power at frequencies overlaps with that of the flash response is increased approximately twice that in darkness before it steadily falls well below the noise in darkness as background light levels increase now you can see the same behavior for foveal and central cones this increased noise at dimmer background says attributed to increased Poisson noise in photon absorption and prior to significant adaptation in the light response so if you look at this cellular noise seems to be more or less similar across location so to relate our signal to noise we actually calculate go ahead and predict cone detection thresholds across background at each of these locations so this is the intensity at which we can detect the response the way we calculate this is as the flash strength or the intensity of light necessary to produce a response whose power matches that of the noise spectrum between 3 and 600 hertz now across background you can see that this detection threshold is quite similar at each of the three locations but there is a slight difference between at the highest background luminance between the foveal and the peripheral cones and we're trying to understand what is causing this difference is it just the noise properties itself or is it a combination of the fact that the dim flash responses could also be slightly different in the foveal location at these higher background lumens all right so there are differences between in both signal as well as noise in foveal versus peripheral cones but if you think about the dominant neural circuit in the fovea which is the foveal midget pathway it has been very well characterized that you know the it has a very private line of communication and luminance adaptation itself in the primate retina has two different sites of operation one is at the level of the photoreceptor and then the other one is at the level of the circuit and in the peripheral primate retina it has been shown that these sites of adaptation switch between themselves based on the ambient light levels so a beautiful study by Felice Dunn and Fred Ricci showed that this at dim light background luminance the cone photoreceptors hardly adapt whereas if you look at the midget ganglion cell in the peripheral retina which is pooling input from about 10 to 30 cone photoreceptors exhibits significant adaptations so if you look at this red curve over here you can see at the dim background luminance you already have a significant decrease in gain whereas you don't see that at the level of the cone or the bipolar cell but as the light level increases to a higher background luminance you see that the cone photoreceptors also decrease in their gain which is then also seen at the level of the downstream circuitry now this is very well represented across a range of background luminance in this summary plot and you can see practically the same thing the solid fit is the cone where you can see very little adaptation going from darkness to this dim background luminance but a great deal of adaptation at the level of the ganglion cell which is shown in these open squares but and then as you hit the higher background luminance which is at this point you see both all the cones adapt and as a result the downstream circuitry also adapt now this was this circuit adaptation has been attributed to convergence or the pooling of cone signals so at dim light conditions when you have very little signal in the photoreceptors in the cone photoreceptors there is a risk of adapting because the signal is so small that you don't want to adapt and that's why adaptation benefits from pooling later in the circuitry at the level of the ganglion cell whereas when the light levels increase the signals become significantly larger than the background noise cones start adapting and as a result the site of adaptation switches from circuit to the photoreceptor so if this circuit adaptation is driven by convergence or by pooling of cone inputs it should be sort of absent in this specialized circuitry where you have no averaging of cone signals from the cone out to the midget ganglion cell in the phobia it's a private line of communication right so in other words cone adaptation should pretty much drive the circuit adaptation or the midget ganglion cell adaptation so we probe this adaptation again plot gain as a function of background luminance and compare them between cones and the midget ganglion cells and you can see the gains the change in gain across background luminance is perfectly overlaid on the midget ganglion cell adaptation so there is no additional circuit component of adaptation at these in this circuitry in the dominant neural circuit and in fact cone adaptation does dominate luminance adaptation in the midget pathway in the phobia all right so I want to briefly summarize what I've told you I've given you many pieces of information but I just wanted to briefly summarize the main take-homes so I've shown you that very in the very beginning that the cone kinetics change in a graded fashion as you go away from the center of the phobia out to the periphery kind of very similar to the how the flick of fusion frequency changes as a function of location or a function of eccentricity and these differences in cone kinetics persist over a wide range of luminance and luminance adaptation in cones if you're looking at the area under the curve the integrated responses they do differ across locations phobia cones exhibit weaker adaptation compared to the peripheral cones which exhibit stronger adaptation and if you look at the from a circuit point of view and in this case the dominant neural circuit in the phobia the midget pathway adaptation in the cone sort of dictate or drive light adaptation in this pathway because there is no pooling of cone signals because of its private line of communication and this regional variation of you know auto receptor and circuit function is also been very well characterized now in other systems for instance beautiful work from the bottom lab on zebrafish where they've identified this strike zone where you have these beautiful specializations of the uv cones which have different properties both at the photoreceptor level as well as downstream circuit level and compared to other parts of the retina and this has also been appreciated and you know well studied in the fly system there there was a in the dorsal frontal region of the male version of these insects there is the inner photoreceptors have much higher you know time course of responses to help them you know achieve higher motion sensitivity when they're in pursuit of females so these regional variations in you know photoreceptor and circuit functions have been very well characterized and evolving a phenomena that is being observed in other model systems as well okay so talked about cone mediated signaling and have talked about how functional properties of cone mediated signaling changes across space and we are able to do that as physiologists because there is a you know immense amount of beautiful anatomical work done by several you know leading anatomists who have sort of characterized how the synaptic connectivity and the circuitry changes take for instance the midget pathway you know beautiful work from Helga Cole behind Swastla Peter Sterling has characterized how that pathway at the level of the cone to cone by midget bipolar midget bipolar to the synaptic circuitry changes but we don't have that level of understanding for the dim light pathway so you know even at the level of the very first synapse between the rods and the rod bipolar cells and you know as I showed you you know when you look at just the density map of the photoreceptors they're not the rod density changes dramatically as you go from the center of the phobia out to the periphery so in the second part or the remaining you know 10 minutes or 5 to 10 minutes that I have I'm going to show you an anatomical analysis where we're going to look at the synaptic connectivity and organization of the first synapse of the visual pathway at different locations and identify if there are any differences in their synaptic ultra structure and this builds on previous anatomical work that has been done you know at this of the synapse to kind of characterize some of these differences across eccentricity and across species as well all right so again you know if you look at the density this you know you can see that the rods are completely absent from the center of the phobia about what degree but they soon start populating you know quite adjacent to the phobia and their density increases dramatically and peaks at about you know 4mm from the center of the phobia and then it again starts falling and this is mimicked in some sense by the rod bipolar cell density where which also peaks at around you know 2 to 4mm from the phobia center so given these changes in both the photoreceptor and the second order neuron how do their synaptic connectivity change so we're going to ask two questions or we're going to look at two features of this synapse the very first visual synapse this is a more updated density map across eccentricity from a recent review by Ulika Grunert and Paul Maunen and we're going to look at the stereotypic organization of the photoreceptor synapse which several of you know is a triad synapse because the pre-synaptic element is opposed to three post-synaptic elements two lateral elements which are offered by the horizontal cell processes and a central element which is formed by the invaginating dendrite of the rod bipolar cell so we're going to compare how does this triad synaptic arrangement change across different locations and secondly we're going to look at how the convergence which means that the number of rods that impinge on a given second order neuron the rod bipolar cell change as you go from you know from central to peripheral retina and we're going to also look at the divergence that is how many output processes are contacted by a single rod axon terminal so to do this we're going to look at four distinct locations with increasing distance from the center of the fovea so we're going to look at the paraphobia which is about one millimeter from the foveal center in fact that's also where the rods are morphologically quite specialized and then we're going to look at central retina which is where the rod density peaks and then we're going to move towards mid peripheral about five to six millimeter and then finally into peripheral retina and the way we're going to do this is by using serial electron microscopy you know using one of the very heavily used techniques to map neural circuits the serial blockface scanning electron microscope I'm going to briefly describe this technique over here we prepare a tissue block retinal tissue block and then how we scan with an electron beam to construct an image and then this is then this tissue block is then advanced with a diamond knife which cuts a very super thin section and then you scan the surface again and this cyclical process of cutting and imaging goes on till you have a 3D volume stack which you then register and then how comes the laborious process of manually segmenting the synaptic structures of your interest and then three dimensionally reconstructing the neural circuits now none of this would have been possible without the you know the instrumental effort by Bernalini Hoon one of my strongest collaborator on campus who single-handedly established the serial blockface scanning electron microscope in our campus as a shared facility and has also helped us figure out how to analyze and sort of get good data out of these projects and also a big thank you to my group of graduate students, undergrads and research interns who have really done this laborious job of manually segmenting this circuitry at four different locations Okay So the very first thing that we look at in these volumes 3D volumes is the ribbons analysis in the rods variable right So here I'm showing you three different examples of rods variable at different locations but have and I've painted in the ribbon which is opposed to lateral elements formed by the horizontal cell processes and the central element by the rod bipolar cell dendrite Now here you can see there are three different configurations so this so any time ribbon is opposed to it's a set of lateral elements and a central element we define that as a synaptic unit Now there are often cases when you have ribbon the same ribbon on either side of the ribbon it has its own set of lateral and central elements sometimes the lateral element is also shared but it has you can clearly see two groups of you know lateral elements and central elements which are feeding into different parts of the ribbons So in that case we define that as a single ribbon and two synaptic unit and in a lot of cases also we see instances where you have two separate ribbons and then each one of them have their again their own group of lateral elements and central element again we define that as two ribbons and two synaptic units So this will be better clear in the 3D rendering where you can see in the first scenario you have a single ribbon and the previous example I showed you that their own set of lateral elements and central elements from their own synaptic unit Now you can see in the middle panel it's the same ribbon which has you know lateral elements and central elements on either side So we define that as a single ribbon two synaptic units and you can see on the rightmost example they're two distinct ribbons Alright, so if you compare the number of ribbons for a spherio at each of these four locations you see something very interesting You see that in the paraphovial rods you have the most instances with ribbons that with two or three ribbons sometimes So the way so the way I've plotted this we've plotted this is the diameter of the circle the size of the circle is proportional to the number of instances of you know one, two, or three ribbons So in the paraphovial rod most of the times you see spherio with or the rod acts on terminal with two ribbons then you see instances of one ribbon and in one or two cases you see instances where you have three ribbons and this is you know a case where paraphovial rods have the highest number of overall ribbons followed by peripheral and then central and mid peripheral Now what about number of synaptic units for ribbon You see that's lowest for the paraphovial rods and almost equal for some of the other locations So it sort of balances out So the number of synaptic units for ribbon versus number of ribbons in the paraphovial Okay so the next we want to ask what about convergence So here I'm showing you on the left 3D rendering of a rod by polar cell in the paraphovial location with all its rod partners and what you can see when we look at the number of rods that converge onto a or make synaptic contacts onto a single rod by polar cell at each of these locations you can see again differences as you go away from the center of the phobia but it's not a linear relationship So you see practically no difference as you go from paraphobia to central retina but then you start increasing as you go towards mid-paraphory to paraphore retina What about divergence So here again I'm showing an example of a paraphovial rod with all its postsynaptic processes and so in this case you have close to six postsynaptic processes three of them are rod by polar cells and three of them are horizontal cell and when you quantify the total number of projections was synaptic projections again you see the biggest divergence at the level of the paraphomial rods and then the paraphore rods and then you have the central and the mid-paraphore rods and when you break them up into rod by polar cell projections and horizontal cell projections you kind of see the same thing as you see for the overall total projections where the paraphore rods have the highest degree of divergence to individual rod by polar to rod by polar cells as well as to the horizontal cell projections All right, so take-homes the highest number of ribbons are present in the paraphomial rods that is somehow being balanced by the less instance of multiple synaptic units Convergence of rod by polar cells increases but increases after the central location highest in the periphery followed by mid periphery and then paraphomial and central have almost the same number of rods converging onto the rod by polar cell and divergences of a single rod to by polar cells or horizontal cell is highest in the paraphomial rods And the big question is what does this all mean for functional implications? What does this mean for synaptic transmission? What does this mean for dim light sensitivity that is all work in progress? So stay tuned Hopefully in the next several months we'll have an answer to that All right, so with that I would like to end and thank wonderful group of students who have contributed to this project Ayns Rosa was a fabulous graduate student Kainath one as an undergrad one as a research intern who have all sort of contributed to the manual segmentation and sort of you know found these interesting differences in synaptic connectivity and organization across different locations SUPIN has contributed to a lot of the analysis in the first part of the talk Jacob who helped talk about some of the data of the first part Fred with a lot of the intellectual input in the first part of the talk Rachel as well with some of the EM of course, Manali Hoon who sort of without whom we wouldn't have been able to do the any of the project and my funding agency and thank you all for your time and the forward questions Thank you very much Ronak for this fascinating talk lots of experiments at different eccentricities and I guess these experiments are not easy to do like the BATS camp ones together with the EM datasets that you showed us towards the end let me before starting moderating this conversation let me go ahead and post this room room link in case people want to join us already and I will start with the first question that appears in the chat it's a technical one from Cyril Eleftheriu are the cones pharmacologically isolated from the rest of the retina during these experiments it refers to the first part of your talk Yeah, so these are all in whole mount preparation so these are not associated so they are you know so they are in a flat mount configuration where we isolate mount pieces of the retina sometimes a big piece which has the phobia and we are they have their their circuitry and their or their connectivity is very well maintained Okay, thank you very much for clarifying and he thanks you he thanks you for answering the question one question I have personally it's like when it comes to the light adaptation like whether it is like photoreceptor or a circuit effect and going like to like it brought to mind the studies that Dowling was doing in the 70s with rays that they only have rod retinas and I was wondering to what extent do you think the absence of rods in the central regions might affect the circuit adaptation Yeah, I mean that's a great question it could you know and to a certain extent we see that the powerful cones you know are adapting more strongly at higher background I don't know to what extent does rods have to play a role in it they could be involved but you know this idea of photoreceptor versus circuit adaptation and in the phobia and in the midget pathway that is something that was expected because we found previously that one of the big gain control mechanisms in the midget circuitry which is inhibition synaptic inhibition is actually very less prominent in the midget ganglion cell so it doesn't get a ton of inhibition synaptic inhibition not nearly as much as a powerful midget ganglion cell and you think of synaptic inhibition and sort of as a gain control mechanism right so so in some sense I mean I'm not saying that's the only gain control mechanism there are a bunch of others but you have you've done away with you know one of the major ones so as a result you know you're not seeing you know that component of circuit adaptation and as a result you have cone adaptation sort of dominate but to your question about whether you know some of this can be mediated by rods yes at the you know we don't fully understand you know when we look at compare cone versus peripheral cone adaptation that there could be a a small rod component to it although I would like to say that we use light levels at which they are saturated but they do at some point start getting relieved right right good point before I return to the questions appearing the chat speaking of rods defining that you have like at the paraffial paraffial rods very limited multiple unit instances this fits well with what we see also for cones that it's kind of dedicated circuitry right yeah yeah yeah they also have synaptic units you can have one ribbon two synaptic units and I think there are some instances also in cones where you know a single ribbon can have two synaptic unit with having the one of the lateral elements shared okay thank you going back to the question from Cyril there's a follow-up so there is no gap junction block either so so we did that experiment where we actually you know what we did was we put gtp gamma s into the patch pipette what it does that it basically blocks the intrinsic response from the cone that you're recording from and what it keeps is the anything that's coming from the electrically coupled neighboring cones and when you do that it doesn't really affect that component is very small you know and and we don't think that has a major contribution towards giving rise to this difference but there is a little bit of gap junctional coupling and that also is light dependent as has been shown in other systems but to what extent that is light dependent in the bovia we don't know or in the periphery we haven't done those experiments right and the other question appearing in the chat currently is from leon lagnado flash response of foveal cone seemed to show a larger undersuit than peripheral cones any implications yeah so that that could be you know because of you know what my dough has found and I don't know I'm speculating at this point so it could be because they have these you know there could be some potassium channels which get activated at those depolarizing potentials which could contribute to that I don't have a good answer to that that's yeah that is interesting but and they could also be a component from the rods although again I would like to say that the rods are saturated at that point but yeah there could be a voltage gated potassium channel there is a difference in their conductance between foveal and peripheral cones right and I have a follow up on that question thanks Leon for that and in case there are no more questions appearing I will be terminating the live transmission after you address my question and we will be continuing offline with people that might be interested so I'm going to post the link once again so make sure you join us after this so the question I have is like maybe it's super naive borderline stupid but because you talk about the time to pick and you also mentioned the integration at some point could it be that also the rise plays a role in the sense of you know maybe at the central retina you depend like strictly to the rise while at the peripheral you kind of integrate more and you wait for the peak itself yeah so that's a great question so what we have done is we have actually normalized it on the x-axis so what you do is so if we normalize it to the peak you can ask whether it's the rise or the recovery and whether one is contributing more or the other and when you do that it kind of falls right on top of each other the foveal and peripheral cone so it's I think it's both you know so in that sense it's both activation as well as the inactivation and that's what we initially thought it was something like rod and cone differences where you could have you still have differences between activation and activation but to a different degree but in this case it's similar so both activation and inactivation is contributing right okay and given that there are no more questions appearing in the chat I would like to thank you once again Ronak for this amazing talk and there are a lot of thank you messages in the chat that you cannot see currently so yeah I'm communicating this as well and with that I will be