 Great session. So we will have a keynote. If anyone has questions for our speakers, we will probably reduce the discussion time. So we do have time for a break. And we'll have the rapid fire and make sure that Dr. Lang has time for his talk and questions. So you can email the speakers with your questions. Thank you. So again, we thank the Knights Templar Eye Foundation for support of our keynote lecture series. And I'm very honored to introduce to you Dr. Richard Lang. Dr. Lang received his PhD from the University of Melbourne and the Ludwig Institute for Cancer Research. He had postdoctoral training with Dr. Michael Bishop at the University of California, San Francisco, and has had a strong academic career working in New York and now at the Cincinnati Children's Hospital Medical Center, where he also holds the Emma and Irving Goldman Scholar Endowed Chair. His scientific interests include early eye development, the role of myeloid cells in development, and effects on the vasculature of the developing R.I. And he's won numerous awards. He's earned them, including research to prevent blindness, Neum Wasserman, Merritt Award, and the Secular Lecture. So I want to have you join me. I'm so excited to hear Dr. Lang's talk. And it will be on Opsons 4 and 5 and Vascular Development in the Eye and the Implications of Therapy. Thank you. Well, first of all, thank you very much, Emmy and everybody else, for the invitation to come and speak. It's been especially interesting for me to hear the clinical perspective on the problems that I'm interested in. I don't often get a chance to come and listen to these kinds of presentations. And so you'll find that what I'm going to present is a little bit different. But we'll see whether it raises some interesting discussion points for the clinical operations. So this is just a reminder about where we live. We live on a ball of rocks suspended in space. There's a star close by. And what that means is that we are washed with electromagnetic radiation with this rhythmic pattern. Now, the availability of photons has been much to interesting a phenomenon for evolution to ignore. And so of course, we have evolved the visual system in response. We've evolved a circadian system that generates a rhythmic physiology. Our circadian clocks are photoentrained. We've also developed what's a much less well-characterized seasonal clock. But what the Lang Lab has been interested in the last few years is the question of whether there are other light decoding mechanisms that affect developmental processes. And that's what I'm going to tell you about today. So from this very zoomed out view to this very zoomed in view, this is, of course, a structure that we all care about and one that's evolved in response to the availability of photons. The Lang Lab for many years has been interested in the mechanisms by which these fetal vascular structures, the pupae membrane, the tunic vascular alentus, and the vassal hyaluridae appropriate or hyalurid vessels, undergo a scheduled regression. It's a very interesting developmental timing problem for us. And of course, we view this as an adaptive process because these structures are light scattering. If the regression did not take place, that would have a negative impact on high acuity vision. And over the years, we've learned quite a lot about the nuts and bolts of the regression process. We know the myeloid cells are essential. We know the wind pathways involved. We know the angiopoietin and wind pathways are integrated in a way that's rather unique. Most of the studies we performed actually on the high load vessel system, as I'll show you. But a little while ago, we became interested. You don't know that this is my best side, right? So a little while ago, we became very interested in the possibility that a light response pathway might be involved in triggering the regression process. I would often get asked the question, how does this process begin? And I had no answers. But we thought about this very carefully and considered the possibility that a light response pathway might be involved. After all, it's an eye. It's a structure that's designed to respond to light. And what we imagined was that when the mouse was born, that was our primary model, the light level would go up. That would initiate a signaling response. And that would be a great way to time regression of these structures. And that turned out to be a very good thought. But it's a lot more complicated than we first imagined. And so this is what we published a few years back now that explains an initial study on how light response pathways regulate vascular development in the eye. And this pathway basically involves this opsin, melanopsin, opium-4 that we know has a number of different functions in the eye. It includes it's involved in circadian entrainment of the suprachiasmatic circadian clock, the suprachiasmatic nucleus circadian clock. It's involved in the pupary light response. This is just another thing that it does. And so through a process we really don't understand very much yet, stimulation of melanopsin at a certain stage of development actually controls retinal neuron fate. And it actually controls the number of retinal neurons that arise. So what I'm going to tell you is basically that an ROP baby has too many retinal neurons. That's what you observe in the mouse model. If this opsin, melanopsin, is a loss of function or you dark rear an animal from late gestation to about a week after birth, you end up with a retina that has too many neurons in it. Cellularity is high. That elevates oxygen demand. You then get elevated levels of VEGFA. Some of that VEGFA remains within the retina. And you get this kind of promiscuous angiogenesis pattern. This is a dark reared animal. And this is P8. Some of the VEGFA makes its way into the vitreous and suppresses regression of the high load vessels. So this was consistent with this whole idea that a light response pathway was important for regulating vascular development in the eye. The real surprise from this project came from data suggesting that the crucial light response with window was not after birth as we'd originally imagined, but it was actually before birth. So the data suggested that the crucial light response window was E16 to E17. In other words, photons are traveling through the body wall of the mother, stimulating melanopsin in the fetal eye, and eliciting all of these developmental responses. Now, some of my colleagues have an issue with the idea that photons can travel through tissue. And so I'm going to start with some complicated chemogenetic rescue experiments that address this question that are a bit complicated, but bear with me. So the first thing you need to know is that opcins are of course G-coupled receptors. And melanopsin actually couples through a molecule called GQ. So if you can do a gain of function GQ, you can mimic light signaling via melanopsin. So one of the first experiments we did was ask whether if you delete, if you do a loss of function of GQ in melanopsin-expressing cells, do you get the same phenotype as a melanopsin loss of function? And the answer is that you do. And so these are just counts of high-load vessels in various genotypes of animal. And if you do a, this is a flux conditional deletion of GNAQ using opsin-faucre, in other words, restricted to the melanopsin-expressing cells. If you do that, you get the persistent high-load vessel network that's characteristic of dark rearing or melanopsin loss of function. So it's just a way of saying that these cells couple through GQ. It's consistent with that pathway. So having established that, we then had the opportunity to take advantage of these fantastic tools that were developed by Brian Roth, the dreads, or the designer receptors exclusively activated by designer drugs. And these are basically engineered GPCRs that allow you to use this unnatural ligand called CNO to activate a particular pathway. And so there's a molecule called HM3DQ that is engineered to activate GQ that is available as one of these stop-flux alleles. So in other words, you can transcriptionally activate this in a cell of choice. And so we decided to do an experiment where we would activate this HM3DQ stop-flux alleles using melanopsin CRE. And so of course, that restricts the transcription activation to the melanopsin retinal ganglion cells. And then we could deliver CNO as a means of activating that pathway in the absence of either melanopsin or the absence of light. And so it's a rescue experiment. We're asking whether we can replace the activity of melanopsin or the activity of light with this gain of function molecule. And I'll show you that you can do that. But the key point about the experimental design is that you deliver CNO before birth. So you're asking whether activation of this pathway before birth is sufficient to elicit all of these developmental responses like discussed. And so you also arrange the genetics so only the fetal mice can respond and only a proportion of the fetal mice can respond so the experiment can be well controlled. And so this is just an experiment where we've done all of that. Here is a persistent high load vessel network in a control animal, but in an animal that has opium-4-CRE activation of HM3DQ has CNO delivered before birth. And it's also opium-4 loss of function. You actually see that you can restore the high load vessel number to normal. So this is an indication that you can replace melanopsin activity with this gain of function HM3DQ but only if you activate before birth. The crucial experiment in my view is this one where you can actually simply withdraw the environmental photons. Normally what you get is this persistent high load vessel network, but if you do this gain of function GQ with CNO injected before birth, you can completely restore the high load vessel number to the normal level. And so this is a nice chemogenetic rescue proof that activation of this melanopsin response pathway before birth is sufficient to give you all of these developmental processes. Now that's a very long-winded complicated experiment. So fortunately, I'm working with this genuine David Copenhagen, who's at UCSF, and he's an electrophysiologist. So he can do experiments like I'll show you on the next few slides where he simply takes a fetal mouse retina and asks whether he can elicit a light dependent response from melanopsin positive cells. And this is one example of that kind of experiment. So this is what we've done here is put a calcium reporter, G-CAMP3, into melanopsin RGCs with opium-4-CRE. This is another stop-fox allele. And then we've stimulated those cells with the appropriate wavelength of light and asked whether we can get a calcium response. And you can, as long as melanopsin is intact. If you do this in a melanopsin loss of function animal, nothing happens. And so this is an E16.5 mouse retina that is showing a melanopsin dependent response. And this is a more recent experiment from David where he's actually done patch clamping for melanopsin RGCs but used different photon flux to try and elicit a response. And so what you can see here is that the latency of the spiking response is longer at lower photon flux. But what's exciting about this is that if you measure the photon flux within the visceral cavity of a pregnant female mouse, you get about this many photons, about 3 by 10 to the 12 photons per centimeter squared per second. And so what David's shown very nicely here is that you can elicit a melanopsin response out of physiological photon flux. And so if you put that complicated chemogenetic proof together with this, you can pretty much be certain now that the retina of fetal mice is light responsive in utero. So obviously one of the big questions, so we're just to summarize there, obviously what we've shown is that light response could be a normal developmental cue and that a developing mammal can respond to light signals in utero. But one of the big questions we face is whether this pathway exists in humans. And obviously that's a bit more complicated to address, but one of the first questions we had to try and answer was what was the equivalent stage of human fetal development to an E16 mouse? One of the ways to do that is just to take advantage of the Carnegie staging that compares surface anatomy of different species. And if you use this scale to make the comparison, what you come up with is that an E16 mouse that has a light responsive retina is about equivalent to a day 58 human fetus. Okay, so that's the middle of a first trimester. So what we're claiming here is the first trimester human fetus retina is going to be directly light responsive. And of course, you've heard all about retinopathy of prematurity. This was the obvious disease in which to address this question because many of the vascular changes you see in retinopathy of prematurity are similar to those you see in a menalopsin null or a dark-reared animal. And so the way we went about doing this, or perhaps more to the point, my colleague Michael Young from CCHMC went about doing this was a multiple logistic regression analysis in which we simply asked the question of whether there was any association between average day length over particular periods of gestation and the risk of severe retinopathy of prematurity in his particular pool of patients. He gathered five years of patient data in which he was able to do this kind of analysis. So obviously what we're doing here is using average day length as a surrogate for light exposure, okay? And this is just a reminder that we have seasons and the day length varies at different times of the year. And so this was a natural experiment. We had kids who were gestated at different times of the year and we could ask whether there was any association between the average day length and their risk of ROP. And so what we're really asking is whether there's a season of gestation dependent risk of the disease, okay? And so this is one concluding table from that study and there are various risk factors that we identified, but here is an interesting outcome that average day length over the first 105 days of gestation after conception, in other words, approximately the first trimester, you get a significant association between that and the risk of severe retinopathy of prematurity. This is one of my favorite data comparisons and this is the chart I showed you originally which is the chart that allows you to approximate the stage of human development that is equivalent to an E16 mouse based on the Carnegie staging. On the right hand side here, I'm showing you a chart that actually has, is very dense with data, but it's one that Michael produced where he's actually assessed the value of his disease risk model for different windows of time from conception, okay? So he's looking, for example, from one to 30 days, 31 to 61 days, 61 to 90 and so on. And the reason he's done this, of course, is that this represents different trimesters and it represents different phases of this developmental process. And so the area under the receiver operator curve tells you about the value of the model and so the higher it is, the better the model of predicting disease risk. And this black line down here is the p-value associated with that model. And what you can see here is that for these intermediate intervals, you have a very significant p-value, but as soon as you flip to the second trimester, which is this 121 to 150 time interval, you lose significance. And so this is a good day because it's the anatomy and the math telling us exactly the same thing. And this is just another way to look at these data. What Michael's done here is document in three month intervals the presentation of a series of patients that need to be assessed for ROP, but then he's represented in the same three month intervals those patients that actually go on to receive treatment. And what you can see is that they're distributed differentially throughout the year. In other words, there is a season of gestation dependent risk of the disease. And so what we can conclude from that is probably that this pathway does exist in humans. And so what we're saying here is that if you have a kid whose first trimester gestation was in the short months of the year, their melanopsin pathway was insufficiently stimulated. They then grow a retina that has too many neurons in it and they are much more susceptible to hypoxia as a consequence. Okay. Okay, so you think that would be complicated enough, but it turns out that there's another option involved in this process. So what I'll tell you about for the second part of the talk here is how this molecule, Opsin V neuropsin, is involved in regulating vascular development of the eye. And this is all about the hyoid vessels. This has very little impact on the retinal vascular as far as we can tell. Direct impact on the retinal vascular. So neuropsin, this is a 380 nanometer violet light sensitive Opsin. It is actually expressed within retinal ganglion cells of the eye, as I'll show you a bit, it's expressed in a whole lot of other places as well. And it raises this really interesting question, which turns out to be true, about whether tissues outside the eye are light responsive due to the expression of this Opsin. I won't talk about that today, obviously. But it has all the features of a classical Opsin. It's been shown to mediate light responses in a couple of settings. And what I'll mention is that Ethan Buren, Russ Van Gelder, have shown that Opsin V actually runs the circadian clock within the retina. The retina has its own independent photorentrainable circadian clock. Opsin V is the crucial detector for that activity. And so obviously, given our interest, we were curious about the possibility that the Opsin V null mouse might have a vascular defect, and it does. This is a retinal ganglion cell expressing Opsin V according to Opium V pre-brainbow activation. These cells are a distinct subpopulation of retinal ganglion cells from the melanopsin-expressing RGCs. And this is the first account of the vascular phenotype in these animals. These are highlight vessel preparations that we generate as flat mounts, and we simply count vessels as a means of understanding what its status is. And you can see that at P1, control the Opsin V null mouse have normal numbers of highlight vessels. But at P8, and this is a stage we typically assess to understand whether there's a persistence or some other phenotype, you actually see that the Opsin V null mouse has fewer vessels than normal. In other words, it has a precocious highlight vessel regression. And this is a unique phenotype. All the other animals that have a highlight vessel phenotype have a persistence response. And so this rather suggested that the mechanism of involvement of Opsin V is going to be different from that of Opsin IV. This is just a nice way to represent the distinction between the Opsin IV and the Opsin V phenotype. Here's the persistence over a time course of the Opsin IV null phenotype, and here's the precocious regression of the Opsin V null. We of course wanted to assess whether this was a light response pathway. And so what we do these days to create light in our mouse rooms is use an array of light emitting diodes, red, green, blue, and violet that target the different Opsins. And they have the same photo and flux as the standard mouse room lights. If you create a lighting conditions that excludes the violet light that open five responds to, you can actually reproduce the precocious highlight vessel regression that is characteristic of the open five null. And so this is an experiment where we actually withdrew the violet light postnatally. And so this makes a pretty strong case that Opsin V modulation of highlight vessel regression is actually a postnatal activity. And so of course this suggests that Opsin IV and Opsin V regulation of this process occurs with distinct timing. And I'll come back to that question at the end. Mintan Nui, the postdoc who was doing the work on this mouse had noted that the open five null mouse had this unusual expression distribution of tyrosine hydroxylase, an enzyme that's involved in the biosynthesis of dopamine. This pathway, we know dopamine has many important functions in the eye, but this pathway is under feedback regulation. And so this disturbance suggested there might be a modulation of dopamine levels within the eye. And that turned out to be the case. And so for example, if you take, if you just do a short time course postnatally and compare dopamine levels within the vitreous of the mouse eye under normal lighting conditions, LD or dark rearing, you get elevated dopamine levels suggesting that light is normally suppressing dopamine levels within the vitreous. And we could also show that in an Opsin V null mouse, you have low levels of dopamine within the neurons of the retina. So this is a lysis to do the dopamine level assessment. But within the vitreous fluid, the Opsin V null mouse has an elevated level. And so that's consistent with this dark rearing experiment. Now the reason this matters for vascular development is the following. And this is a publication from Sinhard, our while back that showed that dopamine can have a direct effect on vascular endothenial cells through dopamine receptor two. And what dopamine receptor two does is activate a phosphatase called SHP2. And SHP2 can defosfolate virge receptor two as a means of inactivating its signaling. And so when we saw these dopamine level data, we wondered whether it was acting as an intermediate and could be an explanation for the precocious regression of the high-level vessels. And so we tested that possibility, or one of the requirements of that hypothesis was that there was a dopamine receptor expressed in the high-level vessels. And here is dopamine receptor two GFP reporter showing your expression in the vascular endothenial cells of the high-level vessels. And this is antibody detection of dopamine receptor two. And indeed, if you do a conditional deletion of a dopamine receptor two, phloxed allele, using PDGFBI-CRE-ERT2, which is a vascular endothenial cell-specific creativity, you eliminate the immunoreactivity with the antibody. So indeed, there's a very clear signature of dopamine receptor two, expression in high-level vascular endothenial cells. This animal with the conditional DRD2 deletion has a persistent high-level vessel network. This is just the count of vessel numbers. And so this is a very clear indication that dopamine has a direct effect on the high-level vessels and its normal function is to promote high-level vessel regression. And so we could also do this kind of experiment where we asked whether on the background of the open five nile mouse, which is this high-level vessel persistence, would the deletion of dopamine receptor two in vascular endothenial cells reverse the consequences? And so you can see that when you do that deletion, that conditional deletion, you end up with a persistent high-level vessel network again. And so this is a clear indication these two have opposing functions within this vascular development pathway. And so one of the predictions of this hypothesis was that the activity of VEGF receptor two within the high-level vessels would be elevated if you didn't have dopamine receptor two present. And so we tested this simply by doing some immunoblots looking at this activating phosphorylation of VEGF receptor two in high-level vessel vascular endothenial cells. And so this is just a short serial dilution of the lysate you can get from both a control high-level vessel and that in which DRD2 has been conditionally deleted. And what you can see is that the level of phosphorylated VEGF receptor two is elevated in the absence of dopamine receptor two and you can quite nicely quantify that here. And so this establishes this proposed mechanism for a direct effect of dopamine in promoting high-level vessel regression. So this is the model that we've come up with that is designed to help us understand how the melanopsin pathway and the neuropson opium-5 pathway integrate to regulate vascular developments in the eye. So this left side here is what I told you at the very beginning where melanopsin light stimulation actually regulates the number of retina neurons that arise and that gives you hypoxia susceptibility, gives you elevated VEGF A if there's either a dark rearing or a loss of function of melanopsin. And on the right hand side here, what we're suggesting is that opium-5 regulates dopamine release to the vitreous and we know quite a lot about the precise mechanisms and I won't go into all of that but it involves this molecule called DAT or dopamine transporter. That's actually regulated by neuropson activity. And so that's that explanation. So then we're suggesting that dopamine that makes its way to the vitreous is actually involved in suppressing the signaling activation of VEGF receptor too and so dopamine has its effect of promoting high-level vessel regression. Now this occurs at a time in development of the mouse eye when dopamine levels are generally rising. So what you get in the absence of opium-5 is a precocious elevation in the level of dopamine. And so we have this opium-4 and opium-5 regulating these opposing influences on high-level vessel regression. And if you work through all those arrows, what you get is that the net effect of opium-5 is actually to sustain the high-level vessels. This is quite interesting. We think this might be an adaptation to get the timing of high-level vessel regression right. Obviously, if it regresses too soon, you're going to create hypoxia in the retina. And so we suspect this is an adaptation to ensure that you don't get high-level vessel regression until the superficial vasculature of the retina is complete. And so that's a way to balance oxygen supply to the retina. And it's actually interesting that there's an interesting clinical correlate. Dopamine is sometimes used to treat hypotension in premature infants. And this has been shown to elevate the risk of ROP. And probably the explanation is that when you give dopamine, you're prematurely promoting regression of a high-level vessel and making the retina more susceptible to hypoxia. So of course, all of this raises this very interesting possibility that you might be able to use these light-response pathways as a way to treat ROP. The first component of this, the Melanopsin pathway, I've told you that we think this is active as early as the first trimester in humans. And so that's going to make it very difficult to take advantage of that in a kid that's born prematurely. Obviously, there's a suggestion that you might take preventative measures. There's a big question now primarily emerging from the circadian biology field that is asking the question of whether living inside under artificial lights is actually detrimental. We obviously evolved outside in sunlight. And so we are under-stimulated arguably. And so there's this interesting idea that maybe our physiology is underserved by light stimulation because of the way we choose to live. And so this is relevant to this question as well. Ensuring that every first trimester pregnant woman gets sufficient light stimulation, that's a massive public health problem. I'm not sure I can solve that one. But it's a very interesting question to address. But this second pathway that we've identified, Opium 5, is a much more interesting prospect because as far as we can tell, it's probably going to be functional in mid to late gestation. And so this interesting suggestion that after a premature birth, maybe violet light wavelengths could be used to suppress dopamine and sustain the high load vessels. Now as far as we know, dopamine doesn't have any impact on the retinal vessels. And so this is a way that you would target sustaining of the high load vessels as a way of suppressing hypoxia in the retina. So that's a really interesting prospect and worthy of further consideration. And so as I mentioned, the effect of dopamine on vascular, we think is restricted to the high load vessels. And it would probably be wise if we chose to do these kinds of studies to deliver light with a circadian rhythmicity because we know there are other issues if you don't do that. And so the sort of experiment where the sort of study we're thinking about now can probably be executed in the next few years because it turns out we're building a new critical care building at Children's Hospital. And I managed to persuade the architects and the people in charge of that project that we should put spectrally tuned lighting in the NICU rooms. And so this is just an architectural rendering of one of the NICU rooms, but the light fittings are going to be research grade, spectrally tuned lighting systems. So in other words, there's a possibility that we could perform this kind of manipulation and ask in a clinical study whether there are any beneficial consequences. So with that, let me finish and identify the people who work with me on this project. This is my fantastic crew. There are a couple of ringers in here. Some of you will notice that this is Russ van Gelder and this is Ethan Buer and Ethan's postdoc, Nico Diaz. They were visiting on this particular day, but I work extensively with Russ and Ethan on all of this work. As many of you will know, they worked on the non-canonical ops since quite extensively. But these are my lab members who've made a contribution and I've had lots of wonderful collaborators on all of this work, including David Copenhagen, Michael Young, Michael Vaughan, Rashmi Hedde and I mentioned Ethan and Nico and Russ and funding from the NIH and the NEI and Urban Goldman Chair. So with that, I will stop and I will be happy to take any questions. Thank you for your attention. And the retina in the, I think the melanopsin knockout, right? So did that, did you also look at how much the vessels extended peripherally and was that increased as well? You do get a slightly accelerated extension of the vessels to the peripheral retina, yes. So that would be useful for ROPs and... Quite possibly, yeah, yeah. It's a wonderful work really and I'm very surprised about the dopamine on the effects of the fetal vasculature. So do you think it has effects on the PFE formation or can be used during the treatment or... It's a very interesting possibility. I mean, persistent fetal vasculature is a slightly different animal from the changes that I'm showing because there are many other cells associated with those complexes. And so it would be simple enough to check and see if the dopamine receptor was present and if it was, then it will become a real prospect, I think, yeah. Great study. Have you done any work in like Scandinavia countries that have such disturbed circadian, they're short days and I mean, that would, you would think it might extend, augment the effects you're seeing and like some place closer to the equator. Yeah, so absolutely. Is there a latitude dependent component of disease risk? And it's actually a question that we are currently addressing with Lois and her collaborators because the data set that they have extends to all parts of the planet. And so we expect to be able to understand whether there's any latitude dependence to the disease risk, but it's a great question. I have a basic question about the vascular biology of the hyeloid plexus versus the intrarational ones. So I want to understand in terms of the oxygen, the impact of vascular perfusion on intrarational arterial oxygen tension of that particular vascular bed. And I ask in particular, I don't know if you remember, but in 2000, you showed that the hyeloid retracts by apoptosis, whereas we then studied the retina and we couldn't find the apoptosis. We found that the vascular endothelial cells actually, when they regress, they migrate into the neighboring vascular segments. So I guess from a vascular biologist point of view, I just want to understand a little bit more about the differences between the biology of the hyeloid plexus versus the intrarational plexus. Right, so I mean, there are clear distinctions in the way the retinal plexi are remodeled versus the hyeloid, that's absolutely clear. The idea that you get a remodeling process that involves cell migration within the retina, absolutely, I think that's pretty solid now. Within the hyeloid vessels, apoptosis is certainly one component of the regression process. Is there a migration component? We haven't addressed that directly. And I think about those two processes as rather distinct. One is a complete regression of the structure. The other one is a remodeling. And to be honest, those sort of nuts and bolts details have not been our primary interest. We've been interested in how you regulate the overall process. So there's still this interesting possibility open, I think. That was so terrific. Thank you very much. We would like to prevent you with a gift. Thank you. From a little bit from Utah. Thank you. Thank you very much. All right, you know, we're gonna do the rapid fire, okay? So yeah, do you wanna, is that okay? So we are gonna have our rapid fire presenters for this session. And what we're gonna ask people to do is come up to either side, present their research for two minutes, and we'll have two-minute discussion questions, okay? So, Markle, are you first?