This symposium is on Advances in the Fundamental Understanding and Therapeutic Applications of Cilia, and our first speaker is Dr. Peter Jackson. So Dr. Jackson is a professor in the Baxter Laboratory in the Department of Microbiology and Immunology and the Department of Pathology at Stanford University School of Medicine. Now unfortunately Dr. Jackson was able to be here to attend the Endocrine Society meeting, but we do have a pre-recorded talk from him, and he will be discussing his work on omega-3 fatty acids, controls ciliary FFAR4 to control adipogenesis. Good morning, my name is Peter Jackson, and I'm sorry I'm not with you in person. I tried to go to Hartford Airport last night to fly to Atlanta, and there were no flights going out, so unfortunately I couldn't even figure out a way to get there this morning. So here I am virtually, and I'd like to tell you about our work, I thank the organizers for the opportunity to speak, and I apologize to everyone for not being there in person. I would really love to be there, but let me jump right in. And I'd like to tell you today about work we've been doing on looking at signaling via primary cilia. And I think as you know, primary cilia have a lot to do with controlling metabolism, obesity, and much of that work has been seen to be connected to feeding control, through hyperphagia and satiety control. But some years ago, we took upon us the idea of trying to figure out whether in the periphery, in fat tissue, in the pancreas, and other important metabolically driven tissues, or metabolically regulating tissues, whether cilia were important, and we find that to be true. And here's a picture of a pancreatic endocrine, endocrine pancreas and the islet cells, and you get the sense that the majority, if not all, of the islet cells are ciliated, the cilia are in red, the nuclei of the islet cells are in blue. And as it turns out, each of the individual islet subtypes, alpha, beta, gamma, delta, each show morphologically distinctive cilia, which tells us time and time again that they have distinct functions, and that's indeed what we find. I want to thank right up front Chen Jingwu, who was the post-doctor who did a lot of this work in pancreas, most of the work, and that was following off work very directly of Karen Hildendorf, who had discovered the role of cilia in adipogenesis. Karen is a new assistant professor at Utah, in the biochemistry department, really wonderful. Chen Jingwu will be starting at UT Southwestern in the virology group in microbiology and immunology, so they're both exciting new labs. We've had a lot of help from Sung Kim, a long-term expert in pancreatic biology, and the founding director of our diabetes center at Stanford. Sung's done a wonderful job. We've been very happy to work with Romina, Yan, Sangbin, and Charles, who've helped out enormously, Janos Demeter, my bioinformatician. We had worked with Daniel Kopinka and Jeremy Reiter on adipogenesis, and Carl Johnson also was a grad student in the lab who worked on adipogenesis. We've been funded by the NIH and NIDDK. I have collaborations with Juvena, Collagen, and Pfizer, but none of them really relate to what we're talking about today. Primary cilia are conserved microtubule-based organelles that amplify signaling in tissues important for feeding and metabolic control, adipogenesis. That's exemplified by critical neurons in the brain, including feeding control or satiety controlling neurons in the hypothalamus, but elsewhere, for example, in the hippocampus. There are olfactory neurons that mediate olfaction, of course, critical for feeding. In the islet cells, there are, as I mentioned, many different types of cilia, both the exocrine and endocrine pancreas, but in the endocrine pancreas in particular, the cilia seem to do important things. In cholangiocytes, the TGR5 bile acid receptor is exquisitely ciliary, and we don't quite understand how it controls bile function. All mesenchymal stem cells, as far as we can tell, are ciliated, including these beautiful pre-adipocytes, where we showed that pre-adipocyte cilia are critical for de novo adipogenesis, but we also recently published with Helen Blau that muscle stem cells require cilia. So there's really an awful lot of roles that cilia can play, both to serve regenerative functions in tissues, but also sensory functions related to the endocrine system. The function of cilia is mediated in no small part by the fact that they're a microtubule-based tissue, they're nucleated from the centroid, and the axonate grows out from that, and the membrane goes around the axonate. And of course, within the cilia, intraflagellic transport, a process of anterograde and registrated transport by kineses and dienes, goes on about five times a second. And that amazing trapping system, we've got an escalator going up and down continuously, is both important for building the cilia and for the signaling process, although there's a lot we don't know about that. Exemplifying that signaling modality, G-protocol receptors are one of the key elements of ciliary biology, and we know many GPCRs now, we've discovered maybe over 40 GPCRs that work with cilia, many, many discovered in my lab. So, how does this all work? Well, why cilia? They're highly localized, amplified signals that work in post-mitotic cells, typically. There's deep evolution from unicellular flagellates that built a highly sensitive adaptable system that can see a single molecule of a single photon. And because, of course, adult disease is often a form of failed adaption to environmental stresses or signals, it's not surprising that as we evolve from single cells to complex organisms, that the problems in the cilia can lead to problems in human disease. This is exemplified here by ciliopathies, a rare set of diseases, thankfully, where Mendelian inheritance of two bad copies of a particular ciliary gene causes a major problem. Retinal degeneration is very common in those who may have a loss of smell, hearing loss, various organ defects. But central in many ciliopathies is obesity, which is related to feeding control. Skeletal abnormalities, reproductive abnormalities, brain abnormalities leading to ataxias and other problems, heart problems, and problems in the respiratory system, kidney, and liver, all of which is related to the failure of cilia. And there are literally hundreds of homozygous mutations that cause a pediatric, generally pediatric disease in these consanguineous populations. But I must note that as we now begin to look, the ability to find many, many GWAS-driven non-Mendelian or complex genetically associated syndromes, effects on obesity and other metabolic problems, we see what looks like a pretty substantial footprint of cilia on ongoing complex human disease. Some years ago, we spent quite a bit of time looking at how hypothalamic feeding was controlled by cilia, and you'll hear from Christian Weiss about this. But here we wanted to focus on the periphery, and we noticed that pre-adipocytes are beautifully ciliated. So in this picture, you can see a phase image of adipocytes, these gigantic fat cells. The blue nuclei are blood vessels that are sneaking through the fat cells. And atop those blood vessels are tiny little fibroblastic cells that have both a cilium and a centriole. And literally, even in a tiny mouse fat pad, there's about a quarter million of these cells, and they actually, we believe, crawl on the cells, on the blood vessels. And we identified that these were the same as the CD34 positive population of cells that have been identified by Rotifer and Friedman. And indeed, we could find that these cells responded quite directly to omega-3 fatty acids and insulin to trigger de novo S phase and the beginnings of fate change that would take these tiny little pre-adipocytes and then begin a long journey over many weeks on a giant adipocyte full of lipid droplets. In order to validate that, we used a pre-adipocyte-specific driver, PDGF receptor alpha, together with Daniel Kopinka and Jeremy Reiter, and we knocked out a key ciliary gene, IFT88, in that fat tissue. And indeed, we could immediately see huge changes in body weight gain over the first 19 weeks of postnatal life when treated with tamoxifen to eliminate the gene around week three. So for females, body mass kind of went south compared to control populations. And if you look at our paper, which was published in Cell in 2019, there really was a huge change on fat mass, no change at all on lean mass. So it really had to do with accumulating fat. And that's exemplified by the fact that both of the fat pads were very tiny compared to control in the knockouts. But also the adipocytes themselves were tiny, suggesting that they had stopped dividing and stopped collecting lipids. We didn't really see any substantial changes in feeding, locomotion, or oxygen consumption. We believe a lot of the excess nutrients, this was on a child diet, but we could see a more extreme effect on a high-fat diet, but that a lot of these nutrients were probably going to be excreted. And that led to a model where these pre-adipocytes I just described to you are sitting atop blood vessels. There are about 40 or 50% of the perivascular cells are these pre-adipocytes. The rest seem to be macrophages, and there may be some interesting crosstalk between that. And in response to these mitogenic signals, like omega-3 fatty acids, like DHA and insulin, these cells can begin to differentiate and begin to make adipocytes. And that, of course, is important as we fatten up for winter, if you're a bear, or if we fatten up for winter because we're eating too much Christmas food. So it's very interesting. We had major good antibodies against FFR4 receptor, which I'll come to in a second. And we started looking at other tissues, including the bile duct and the brain and the pancreas. And we immediately noticed that there were beautiful ciliated cells within the pancreas that express some of these other receptors. As you know, the endocrine pancreas is made up of a series of ductal cells, and then the endocrine pancreas itself, which has alpha cells secreting glucagon, beta cells secreting insulin, gamma cells secreting pancreatic polypeptide, and delta cells secreting somatostatin. And recent work from Alton Gerdes, but especially from Gene Hughes and David Piston, showed that if we knock out the IFT88 in an insulin receptor, insulin driver, in order to specifically in beta cells, that would cause a substantial loss of glucose-stimulated calcium signaling in beta cells, leading to glucose intolerance and diabetes, although without any particular obesity. So it was really driven by the beta cells themselves. And so we were trying to explain that as this work came out. And Chen-Ting Wu began with using single-cell profiling data from Sun Kim's lab to try to identify alpha and beta cell, among other types, selective GPCRs. And we had over the years established a protocol where we would take GPCRs and make a C-terminal fusion to GFP, express them in cells, and then assess whether they were ciliary. And we found quite a few candidates, somewhere on the order of nine at least, that were found in beta cells. And we used first these alpha TC9 alpha cell lines, or these min-6 beta cell lines, just as a starting point. And we were able to show that those cells were highly ciliated, both cell types. And, for example, the omega-3 fatty acid receptor FR4 was expressed in a majority of both alpha cell cilia and beta cell cilia. In contrast, the receptor PTGR4, which is very beautifully ciliary, here you can see that it was not particularly ciliary, nor expressed in the alpha cell, suggesting a degree of beta cell selectivity. We found some other receptors, like the cuspeptin receptors, and adrenergic receptors, and some other known ciliary receptors. And we haven't begun to explain those yet, but they're potentially all very interesting. We were able to then get antibodies to both FR4 and PTGR4. We made antibodies to FR4, and could show and confirm that the endogenous forms of FR4 receptors were ciliary both in min-6 cells, beta cells, and alpha TC9 alpha cells at a very high percentage. And, again, PTGR4 seemed to be selectively expressed in beta cells and not so much alpha cells. Just to describe these two critical receptor signaling systems, omega-3 fatty acids include – receptors include FR1 and FR4. These are the long-chain fatty acid receptors. Omega-3 fatty acids are ancient signaling lipids produced by highly conserved desaturated – desaturases. And they're sourced from dietary fish or nuts, salmon, for example. And they're very important for neural, retinal, and metabolic health. And there's been much published in the sort of nutritional literature suggesting their importance for normalizing diabetes. And DHA and EPA are two of the best understood. They're not actually aromatic. They have double bonds, but they're every third carbon. And these very unusual structures form a very compact form of lipid, which then has a very important signaling properties in the membrane. And, indeed, these two receptors have been deorphanized by the crystal structure of FR1, which showed a beautiful binding pocket for launching fatty acids. As it turns out from the work of others and now us, FR4 signals through a dead light cyclase. It's a classic cyclic MP driver. And we – I'll show you that in a moment. It activates both protein kinase A and the EPAC exchange factor, which activates RAPG TPAs. It's important for membrane insertion of many proteins. In contrast, FR1 is not ciliary. And it works for calcium signaling, which is very important in the pancreas. We have, luckily, very exquisitely selective pharmacological tools to hit FR1, this TUG 891 compound, for example, or FR1, this TUG 424 compound, so we can signal through each receptor individually. We can use the native ligands, and they work, but the pharmacological drugs work a little bit better. The PTGR4 receptors are four in number, and they respond to a particular inflammatory prostanoid or prostaglandin, PGE2. That's the most important prostonite in the islet, but also in many inflamed tissues, for example, the kidney or the brain. And again, two of these receptors, PTGR4 and PTGR2, work through cyclic MP. And actually, cyclic MP agonists can serve as insulin secretagogues, driving insulin secretion in pancreatic beta cells. When knockouts of PTGR2 and 4 were combined, there was a strong enhancement of type 1 diabetes in a particular model. So, we want to understand how this works, and just to describe the system, we take them in six cells, we add agonist, and then we add glucose. And 30 or 60 minutes later, we isolate supernatants, but then we also lyse the cells, and we do a lyse is to look for insulin from the beta cells in both pools. And thus, we can calculate the secreted insulin divided through by the total cellular insulin in order to get the amount of insulin that's secreted. That's a nice, stable assay. And so, we look at full, compared to increase, to control. And you can see here that in control cells, if we start with basal levels of glucose, about 3 millimolar glucose, then there's a small amount of insulin secreted. But as we increase glucose levels to 17 and 25 micromolar, we see a notable increase in glucose secretion, a glucose-dependent insulin secretion. But if we now add agonist to FR4 at different doses, and this is generally dose-dependent, we don't really see much activity in the basal level of secretion. But we see strong enhancements of the glucose-enhanced secretion here at 25 millimolar, for example, going from here to here, and from here to here. We see the same thing with TUG891, even better. These are two different chemical agonists, and also DHA, a native ligand, is the same thing. So, we see stimulation not in basal glucose-stimulated insulin secretion, but in that that's agonized by the receptor itself. And we see something really quite similar with PDGFR4 receptors that are pretty strong in enhancing the red bar compared to control, showing that they stimulate glucose-stimulated insulin secretion rather than basal levels of insulin secretion. And it's interesting to now look at what happens in beta cells. You may know that beta cells respond fairly monotonically to glucose as you increase the levels of glucose. This is our experiment that you see more insulin secreted. But in alpha cells, you see something different. It starts fairly high, the amount of glucagon secreted, but then at higher doses of glucose, the levels of glucagon secretion go down for a while and then back up again. So, it's this biphasic response. And this has a lot to do with normalizing late release of glucagon to maintain glucose levels after insulin has been secreted. You don't want to drop to very low levels. That's bad, of course. And here, what's really interesting is that we see, in contrast, the FFR4 organism shows a strong increase in the basal levels of secretion at low levels of glucose in the blue bar. But less of an effect at the high levels of glucose. And we don't see any effect with PDGR4, suggesting that the fact that it isn't expressed, it probably isn't very important. So, we see a clear signaling response. And I'll show you in a minute that it works in the same pathway. But nonetheless, it's a quite distinctive response. And I will go through all the data. It's in the paper. This was published last year in Genes of Development. We can see, indeed, looking at antibodies in both native human islets from cadaveric donors or from mouse islets, which we get weakly from the diabetes center, that both FFR4 was ciliary and PDGR4 was ciliary, selectively in beta cells, but FFR4 only in alpha cells. And, again, I won't show you, but we can stimulate using those isolated islets. We can use these same drugs to enhance glucose-stimulated insulin or glucagon secretion and see a very strong effect in purified primary mouse or human islets. Just to emphasize that again, here's the basal level secretion. I'm just showing this fold increase. But as we add this FFR4 agonists or PDGR4 agonists, we can stimulate the glucose-stimulated, the high glucose or middle glucose levels of insulin secretion. And, again, FFR4 stimulates glucagon secretion, but PDGR4 agonists do not. So to summarize this first part, the endogenous FFR4 and PDGR4 receptors localized to cilia and beta cells, FFR4 itself localizes also to alpha cells. They have somewhat different outputs. And FFR4 and PDGR4 agonists stimulate insulin secretion in beta cells and only FFR4 agonists promote glucagon secretion in alpha cells, but, again, at basal levels of glucose. And I think as a paradigm, this is the kind of thing that we want to study. We want to understand how individual cilia function in a very local context to receive signals that probably come in from the vasculature or from circulating hormones or metabolites and turn on specific receptors in very localized places in the body, in the brain, in the pancreas, and on other walls. But we want to make sure how that works properly, so we need to look at some mechanisms. And for many years, my lab has been using affinity purification mass spectrometry and other forms of characterization to look at critical receptor trafficking molecules and try to understand the mechanisms of ciliary signaling. And one of the key molecules we found was a molecule called TULP3. It makes a complex with the intraflagellar transport A complex, an important complex for signaling. And it turns out it's also important for trafficking GPCRs into cilia, and so is TULP3. The way TULP3 works is it had one domain that binds, we believe, somehow to ciliary localization signals within receptors. Those are generally in the C-terminus, in the IC4 domain. And also another domain that binds quite beautifully to phospho-inositides, all those specialized phospho-inositides that are lacking the 5' phosphate. Those are removed by a specific phosphatase, INPP5E. And thus, somehow, this trafficking molecule, TULP3, can find a particular zone of the cilium at the base of the cilium and thereby invest GPCR insertion at this point for signaling. And we're working quite hard on trying to understand more details of this mechanism. But we could show that when we knocked out TULP3 by CRISPR knockouts in both Mn6 and Alpha-TC9 cells, that the receptor FR4 lost its ciliary localization, although the cilia were somewhat uninfected in these cells. Again, PTGR4 in TULP3 knockouts did not localize to the cilia, but the cilia were intact. You can see strong effects on the intensity. And again, in Alpha-TC9 cells, FR4 localization to cilia was also lost, a little less dramatically, maybe. So that was consistent. Now we want to know what happens when we're missing this key molecule, TULP3. There are other contexts from mass knockouts that TULP3 knockouts cause a strong effect on ciliary signaling. And so what we can see quite clearly here, looking at basal levels of insulin secretion or glucagon secretion, but here insulin secretion, TULP3 knockouts really didn't do very much. But when we now look at the ability to stimulate insulin secretion, or GSIS, in response to the FR4 agonist, the TULP3 knockout concurrent with losing ciliary localization of the receptor caused a notable decrease in the amount of signaling. And that was true for the FR1 receptor, which doesn't work with cilia. but it was also true for the PDGR4 receptor that works through PGE2 agonists. And again, we can see something pretty similar in glucagon secretion in alpha cells that, again, the FAR4 agonist had a pretty strong effect on, had a strong TELT3-dependent effect on glucagon secretion. I'd like to also note, and this is really fascinating, which is that we consistently see in many of these cell signaling experiments that when we inactivate cilia, or in this case, inactivate TELT3, that the basal levels of secreted factors, in this case insulin, rise pretty substantially. And we're very interested in the role that cilia serve to control basal repression of secretion. If normal secretion is suppressed by cilia, then cilia become critical gates for activating insulin secretion. In the absence of those, presumably the secretion becomes more ungated, and that would lead to hyperinsulinemia or other kinds of misregulations in metabolism, which could be very interesting. And finally, I won't go through this, but we did some pretty heroic work with Yan in Sun's lab to both use human and mouse islets to take the islets, disperse the cells, do knockdowns of TELT3, and then re-aggregate the cells to make what we call pseudo-islets, a kind of organoid-like structure. And I won't go through the data, it's in the paper. But again, we can show TELT3-dependent stimulation of glucose-stimulated insulin secretion in these pseudo-islets. So that was really a good point if you're a replicant to that experiment. It looked surprisingly good. And finally, following the work that Karen Hilgendorf had done in 3D301 cells, we could use a reporter for ciliary cyclic AMP as a reference ciliary marker in red, and then a GFP reporter in green, which binds to cyclic AMP, which reduces fluorescence. And you can see upon controls, you can see the red reference marker and the green cyclic AMP marker are concurrent. But when we add the FR4 agonist, we can see a reduction in cyclic AMP, an increase in cyclic AMP concurrent with reduction in fluorescence. And that's quantified beautifully here. It takes a couple of minutes, it's very fast. And again, we can see something fairly similar in TC9 cells. And so this is an exciting new signal that we can study if we're beginning to try to understand which ciliary cyclases and which signaling molecules are critical for that signal. We also want to try to link these treatments to the clinical treatment for diabetes. I think many of you know that GLP1 or GYP1-like receptor agonists are approved for treating type 1 diabetes in the US. It's one of the biggest drugs in the world. And we wanted to test whether FR4 stimulation, GSIS, similar to what these GLP1 agonists extended and others were in xenotype, and ask whether the two could synergize as a way to understand whether that would make a nice drug combination. And we found that maybe also FR1 agonists would work because those who stimulate calcium will get important for cyclic AMP dependent for release of insulin. So we tested these two in combinations. And indeed, we could see stimulation of GSS relative to control in both FR4 agonists and GLP1 receptor agonists. But we saw strong combinations when we mixed the two together. We can also see strong combinations of FR1 and FR4 agonists mixing together. And I'm not showing all the data here, but all combinations of the three various systems work. And we believe, in part, that this works because FR4 will work through cyclic AMP. GLP1 receptor works through cyclic AMP, and FR4 works through calcium. Those two critical channels of calcium signaling and cyclic AMP signaling conspire to both recruit more vesicles to the membrane in the beta cell and also stimulate fusion of those vesicles to release insulin. So summary two is that FR4 requires TELP3 to localize to cilia and alpha and beta cells, both in vitro and in islets. FR4 stimulation of glucose-stimulated insulin secretion requires TELP3 in cells and in islets. Combinations of FR4, FR1, and GLP1 receptor agonists show notable cooperativity in stimulating insulin secretion. And we think that it may have very therapeutic implications. I didn't show you that the cyclic AMP-dependent activity in beta cells requires both protein kinase A and EPAG signaling. There's good agonists, but that's in the paper. And we are now doing extensive fossil proteomics to determine the specific pathways activated by these agonists in beta cells. And that's really quite fascinating. I will skip this. You should also know that we're doing an awful lot of proteomics on pancreas. And we're finding surprising complexity. Some Kim's lab has been able to individually sort different islets, sometimes alpha, beta, and delta. And we're looking at the individual factors that they secrete and that they express. And it's a very rich and interesting story, which we see developing. And I'd also like to note that my lab started working on COVID in March 2020. And last year, we published an exciting paper, I thought, that showed that we took advantage of our ability to culture human islets. We infected them with SARS-CoV-2, a live virus in a BSL-3 facility. We're able to show that that infection caused a selective infection of beta cells, a loss of insulin secretion, and a loss of beta cell liability. And when we look back at patients who had succumbed to severe COVID, those patients actually had a surprisingly selective beta cell expression of viral antigens. More recently, large epidemiological studies have supported that patients who have severe COVID, whether either in the hospital or actually the ICU, have a substantial increase in de novo diabetes based on 660,000 patients in the AIDS system. So we'll see what happens in the long term, but it's a little scary to think about COVID causing long-term beta cell damage and diabetes, and something we need to really kind of consider. So anyway, I'd like to thank you all for listening. I hail back from Stanford, the home of the primary cilium, that's Hoover Tower, which looks very ciliary. We're quite interested in more people who are interested in endocrinological signaling, not only in the beta cell, but probably in some other tissues that I find interesting. Again, I'd really like to thank Chen-Ting Wu, Karen and Sung, Sung's lab, for all the help and support with that. And with this, I will stop and thank you all for your attention. Again, I apologize for not being there in person. I don't know if I can answer your questions, but please email me at pjackson at stanford.edu, and I'm glad to talk about the work and let you know any answer I can provide. Thanks a lot. Have a great day. Thank you. Okay, so I have the pleasure of introducing our next speaker. So this will be Joan Jorgensen. So she is a professor at the University of Wisconsin School of Veterinary Medicine, and she received her PhD from Case Western under the tutelage of Dr. John Nielsen, and she'll be presenting on hedgehog signaling via the primary cilia and fetal latex cell development. And so without further ado, just welcome our speaker. Okay, now that I know how to work the slides, good morning, everybody, and I'm very excited to be here, and I want to thank all the organizers, Eileen, for inviting me, and I'm happy to see everybody that found the room in the notorious B Hall. But I'm here to tell you a little bit about fetal androgen synthesis, and to start, I thought I'd talk about the adult testosterone levels in the serum, and thinking about a population problem that we have with adult levels of serum testosterone. And indeed, basically, what I'm showing you here is a population problem that we have with adult levels of serum testosterone. And indeed, basically, what I'm showing you here is I'm going to learn how to use the pointer without lasering you two. So what I'm showing you here is some serum testosterone levels that have been shown over different time frames. So here we've got serum testosterone levels over age, and what I'm showing you is in the mid-80s, we had levels that were up here, and the higher levels, and as we've gone up through 15 years, what we observe here is that testosterone has significantly declined. And this hasn't been updated, but it's kind of scary to think what's happening now in 2020. And so this is also partly contributing to the huge increase in testosterone prescriptions given to older men, and to elevate their activity and giving them a positive quality of life. So what's causing this population decline in testosterone? Well, some of the evidence points to that it's actually a fetal problem. So we think, the thinking now is that the testosterone problem actually begins during development. And there have been some investigations, primarily from the Sharpe and Skakebek labs, that have observed a constellation of phenotypes that are related to reproductive tract disorders and thought to be related to actually fetal androgen synthesis. And these will include variations in sex differentiation, including cryptorchidism, which is basically when the testis fails to descend into the scrotum, and hypospadias, where the tip of the urethra is not at the end of the penis, but instead somewhere along the ventral shaft. And so this is where it's first noticed, but then later on in life, these young adults are showing increased incidence of testis germ cell cancers at a relatively young age, and infertility noted to be contributed by low sperm counts. And this, along with the decreasing serum testosterone levels in aging men, are all increasing over time, which has been quite an alarm. And so where is this androgen being produced during fetal development? Well, we know that there's, in mammalian species, and actually all species, we know that there has to be a certain amount of testosterone produced in a very critical window of time. And this is coined the masculinization window. And so that timing is different depending on the species, but we know that the timing is very early in development, and there are certain thresholds of androgens that have to be released. And it's thought that those androgens that are made in the fetal testis are important for secondary sex characteristics to develop, but also are thought to program the adult progenitor cells in the testis. So what is the source of these cells? So what I'm showing you here is a rendition of the testis cord, and basically what that means is early in development, the first cells that will differentiate are these Sertoli cells. And once those Sertoli cells differentiate, they will start to corral the germ cells into cord-like formations that will eventually become the seminiferous tubules in the adult testis. And outside of the cords, there are a number of what we call interstitial cells, so basically cells outside of the cords, that will also be present and are under control of whatever the Sertoli cells produce. And over time, these interstitial cells will include a population of cells called the fetal latic cells, and those fetal latic cells will produce androgens that will then go on to program the other parts of the fetus. So what is this, and how does this link to primary cilia? So we know in order to make androgens, there are certain components that are necessary. We need cholesterol, and I'll get into this in a little bit more, but we need desert hedgehog or hedgehog signaling pathway, and we know that there's some intricate contributions between both cholesterol and desert hedgehog, and these are all sort of combined and related to that primary cilia. In addition, if you have any sort of mutations in cholesterol synthesis, desert hedgehog signaling, or ciliopathies, any failure of any of these components will lead to variations in sex differentiation and oftentimes much more severe developmental problems. So we also think that there's probably other processes that the primary cilia plays in directly regulating androgen synthesis. So I'll start with each piece of these. So first off, cholesterol, this will be obvious to you. Cholesterol is an integral component of the cellular membrane, and what's really interesting is that the cholesterol within the primary cilia itself, the transition zone, and the plasma membrane is very different within those individual spaces. In addition, cholesterol is the substrate of androgen synthesis. So we have cholesterol is the primary substrate that goes through a series of enzymatic conformations to finally make all steroids, including the androgens. So what about desert hedgehog? Well, hedgehog, as you know, is a classic portion that requires the primary cilia for it to function. In addition to the hedgehog pathway, we know, as we just learned from Dr. Jackson, that there are a lot of other signals that go through the primary cilia, including Wnt4 and platelet-derived growth factor alpha, which is also an important mediator of latex cell development. So I'm going to focus on the hedgehog pathway here, and in particular, the desert hedgehog family member, which is almost exclusively functioning in the fetal testis, and others, the testis and then later on the ovary. So here what I'm showing you is the testis cord with the sertoli cells in yellow, and as they differentiate, they will make a desert hedgehog, which will then be secreted and act upon those interstitial cells that are lying right outside that testis cord. Once desert hedgehog binds to its patched receptor, it will allow the activation of smoothen within the primary cilia, and then the downstream GLEE factors, including GLEE 1, 2, and 3. And we know once we have an increase in the activity of the hedgehog pathway, we have an increase in sterogenic factor 1, which is basically the master regulator of the sterogenic enzymes, which then allows an increase in them and androgen synthesis. So a lot of studies have been undertaken to try to sort out which part of the hedgehog pathway is important in the fetal latex cells. And so studies have shown that if you knock out GLEE 1, basically GLEE 1 knockout itself seems to be unnecessary for the entire fetus to develop. But if you knock out GLEE 2, there are a whole lot of issues that occur. But interestingly, fetal latex cell development is not one of them. So the fetal latex cells actually develop normally. Well, as normally as we can tell, they develop androgens in the fetal latex cells. And so Ivar Barsoom in Humphrey Lyle's lab then used a small molecule inhibitor called GANT, which basically binds the DNA binding domain of all GLEEs to block GLEE transcriptional activity to just show that it actually functions through GLEEs. And that's when they were able to recapitulate the loss of the fetal latex cell differentiation. And so that left the poor stepchild of GLEE 3. And so GLEE 3 is always kind of considered later on. It's always thought to be more of a repressor. And in fact, when GLEE 3 is knocked out, more often you get a phenotype of excess hedgehog activity because you knock out the GLEE 3 repressor action. And so we were hypothesizing, well, perhaps GLEE 3 is playing a different role in the fetal testis. And so we started to look at this using a mouse model called the GLEE 3 extra toes J model, which is one of the first mice that basically made Jackson Labs way back in the 50s. The interesting thing, too, is if you look in the human population with GLEE 3 mutations, they have a lot of different phenotypes. But they do include phenotypes related to variations in sex differentiation, including cryptorchidism and hypospadias. And so this particular mutation in mice basically renders GLEE 3 nonfunctional. So it's considered a knockout of GLEE 3 activity. It's also considered a model of this human Grieg cephalopolysyndactyly syndrome, where it's pretty much a functional knockout in the human. So those patients have shown evidence of cryptorchidism and hypospadias. And we were using these mice to study that process. So what I'm showing you here is sort of a timeline in mouse days of the transition from a bipotential gonad to a testis, focusing on testosterone synthesis. When it begins around embryonic day 13 1⁄2, the synthesis peaks around 16 1⁄2. Unfortunately, those embryos die just before birth, with gestation ending right around days 19 or 20. So the first thing we wanted to see, okay, do they recapitulate that phenotype? And what we observed was that there was cryptorchidism in the GLEE 3 mutant mice. We also found evidence of hypospadias in the GLEE 3 mutant external genitalia. And then we looked at, based on those, okay, is it because of lower fetal testosterone? And we measured testosterone from fetal gonads. And at embryonic day 13 1⁄2, we saw that there was some testosterone present, but not as much as the controls. And more importantly, we found that as you reach the peak of testosterone synthesis, that measurement didn't budge in those mutant fetal testis. So where does the fetal testosterone make? It's made in those fetal androgens, or I'm sorry, the fetal latic cells. And so we looked at histology and a lot of different markers of all the cell types in the fetal testis. And I can assure you the Sertoli cells looked like they were functioning. The germ cells were all functioning and present. The testis cords were all there. But when we started to quantify fetal latic cells, we observed that there were fewer of them. But interesting to us, there were some there. So we did see some fetal latic cell differentiation, but there weren't as many cells. And when we evaluated the output of those cells, we noted that their sterogenic potential was significantly decreased and got worse over time. Some of the other things we noticed that I don't have time to show you, but we also noticed that the hedgehog pathway genes were also decreasing over time, suggesting to us that those fetal latic cells were actually losing their identity. So we looked at the primary cilia and wondered if there was an issue with what their function. And what I'm showing you here is some measurements of gene expression of important cilia or pathway, basically genes important for the cilia itself. And we have some images here to show that not only are the primary cilia present on these fetal latic cells with the red stain, but present everywhere in those interstitial cells. So based on that, we thought, well, let's go back to GLE-3 and see if we can give back the full length GLE-3 now and see if we can't sort of rescue some of this phenotype. So unfortunately, there's no way to do this in a cell line because there are no fetal latic cell cell lines. So what we've done is we've developed a method to inject plasmid DNA into fetal gonads ex vivo, and then we electroporate them, and then we monitor expression of these plasmid DNA in the gonads themselves. So the first thing we did is we injected fetal testes. And what I'm showing you here is just a copy number PCR of GLE-3 itself. And what you see here is a wild type testis in purple, and we've injected just a CMV EGFP control plasmid DNA. And the GLE-3 level was expressed here, and that we considered to be basically wild type expression of GLE-3. And what you see here in the orange, well, you can't see, is endogenous GLE-3 with that same EGFP plasmid in the GLE-3 mutant mice. So showing again that the GLE-3 is basically knocked out. And then when we add the GLE-3 full-length plasmid, we were able to rescue GLE-3 expression about 40%. And so when we looked at that, we looked at some negative controls. We know that GLE-2 itself and then SOX9, a Sertoli cell gene, is not regulated by GLE or hedgehog pathways. So they did not change, but when we looked at GLE-1, we did get a significant, almost five-fold increase in GLE-1 expression. So we think even though we've got about a 40% rescue, we were able to functionally rescue the GLE pathway based on GLE1 marker. And when we looked at the sterogenic enzyme pathway, we found that SF1, that master regulator of sterogenic pathway gene expression was increased, and then likewise downstream, the star CYP11A1 and CYP17A1 pathway genes were also increased. So based on this, we're kind of rethinking a little bit how the regulation in the fetal atyc cell goes. Now we have a new organ or pathway where GLE3 activator form is actually essential for partly differentiating the fetal atyc cells, but also maintaining their phenotype, where we have GLE3 actually promoting GLE1 activity, SF1 and other sterogenic enzymes. We know that the GLE pathway genes can act directly on sterogenic enzyme promoters. We have been trying for a long, long time to see whether GLE3 can activate SF1 promoter directly and have failed every single step of the way. There are hedgehog responsive elements within the entire enhancer region of SF1, but when we look at those, none of them have been fruitful. So we think it's an indirect mechanism. And we also know, based on some of the other studies I didn't tell you about, that the GLE3 seems to maintain the maintenance of the fetal atyc cell itself. And so thinking about that, we know that the hedgehog pathway is important. We know cholesterol is important. I wanna show you a little bit more about how intimately these are connected as well and think about how that relates to the primary cilia. And so first off, with the hedgehog pathway itself, we know that hedgehog is made in the Sertoli cells, but in order for it to be functional, it has significant post-translational modifications, including the addition of cholesterol. And this is a covalent modification. And so once hedgehog has palmitate and cholesterol added, it can be released and is an active ligand. In addition, what I think is also super interesting is the idea of cholesterol. So both PATCH, which is a 12 transmembrane type receptor protein, and SMUTHEN, which is a seven transmembrane type protein, both have what we call sterile sensing domains, sterile meaning like cholesterol and other oxysterols, and it can bind to cholesterol. And one idea of how this might work is when the hedgehog ligand is not present, basically the PATCH has a pump, the sterile sensing domain pumps membrane cholesterol away from the SMUTHEN receptor, making it inactive. Once the hedgehog actually binds with the cholesterol binding into the sterile sensing domain, it will block that channel, and it allows that cholesterol to accumulate, and that ciliary cholesterol membrane component changes such that the SMUTHEN, it's much easier for it to actually pump its own cholesterol, but also move around in the cilia to get to the appropriate position to then activate the GLE transcription factors. So then how else can we think about how these all might integrate with each other and include in the primary cilia to directly regulate the androgens? Because as I showed you, at least from some of our data, the desert hedgehog pathway and the GLE transcription factors need something else in order for them to function and make an appropriate level of androgens. So what I'm showing you here is copy number PCR of four different steroid pathway genes. And what we did is we painstakingly took harvested gonads from embryos every four hours, starting at 11 1⁄2 up to embryonic day 14, and then we spaced it out a little bit more to every 12 hours through the masculinization programming window of the fetal testis and the mouse until about the time of birth. And with this, we were able to compare steroid pathway genes to each other with copy number PCR2, which was helpful to us. So the first thing we noticed is by embryonic day 14, what we observed was there was a generalized increase in all these pathway genes, but we felt like it was more speaking to the number of fetal aiding cells that were differentiating. And so we think that this is basically indicative of differentiation by embryonic day 14. You aren't gonna count any more fetal aiding cells, but they will become better at making androgens. And so this is the part we know Desert Hedgehog and GLE3 are important in differentiating these fetal aiding cells, but we're not sure. We think that this is basically just an increase in number of cells. But then what was surprising to us is between 14 and 16, there was a profound increase in steroidogenic enzyme pathway gene expression. And so we know from our data before that based on even just testosterone synthesis, that with the knockout of GLE3, we had no more increase in androgen production. And so we think that there is some component there, but there were fetal aiding cells present. So we think that there has to be something else that's actually stimulating these cells to produce androgens. And then after 16, E16, the expression of these pathway genes dropped like a rock. And that was, you know, a lot of times we see this and we know that fetal aiding cell androgens are decreased and basically gone by the time of birth in mammalian species altogether. There is a little bit of a mini puberty after birth, which is super interesting. And then of course, puberty later on. But this idea of what happens to those fetal aiding cells afterwards has just now begun to be investigated. So now, right now, we're really interested in trying to figure out what is the stimulus here. And we think that whatever this stimulus might be is probably something that can be activated in this time, but then blocked or repressed afterwards to cause that de-differentiation. So where do we go? We think about adult cells and basically any sterogenic cell that we're all familiar with. We know that the adult aiding cell is able to produce testosterone on its own. It has all the machinery present. It does not have any primary cilia. And it also depends on an external signal. So the primary signal occurring from via the hypothalamic pituitary axis with the release of luteinizing hormone that stimulates the LH receptor, a classic GS subunit protein-coupled receptor. It releases a number of pathways, but the primary pathway is to the cyclic AMP and PKA pathway. And once that PKA pathway is released, we know that it phosphorylates star itself, and that's what helps initiate that transport of cholesterol into mitochondria to allow the rest of the pathway to go on. And then it also phosphorylates transcription factors that can bind to the promoters of all of these pathway genes to stimulate stereogenesis itself. So how can we think about this and relate it to the fetal testis? The interesting differences here are that the fetal testes do have primary cilia, and they also cannot make testosterone on their own. So it's a two-step process in the fetal testis. They have all of the machinery to make androstenedione, but then that has to go back to the Sertoli cell where HSD17 beta is made to allow conversion into testosterone. So there's a tight back and forth between the fetal latex cells and the Sertoli cells all throughout fetal testis development. And so the other important thing to think about is this LH receptor. So we know that the hypothalamic pituitary axis is not quite set up. In humans, there is the presence of HCG, but we also know that the LH receptor isn't necessary for fetal stereogenesis. If you knock it out, or in humans that have an LH receptor mutation, they have no trouble making androgens during fetal life. So there are other G protein-coupled receptors, and even GS subunit protein-coupled receptors present in that fetal testis. So we're wondering if cyclic AMP and PKA are even active. And so that's our next step, is try to understand whether this PKA pathway is active. And so a little bit of preliminary data I can show you today is that we've taken out fetal testis, we've put them in a hanging drop, basically ex vivo culture. We've treated them with bromocyclic AMP, and we've done this with LH and HCG, and we know we can induce the stereogenic pathway genes. But I think the key indicator is we've used this other small molecule called HG9, which is basically a molecule that blocks the blocker of the cyclic AMP pathway, which is the salt-inducible kinase pathway. And so what I'm showing you here is just some quantitative PCR. We know that the cyclic AMP has no impact on the SF1 expression level, but with a bromocyclic AMP and this HG9, we get an increase in CYP17A1, which we also know is not that profoundly impacted by the PKA pathway. However, star is a classic one that is quite significantly induced, and what we're most convinced by, not surprising that cyclic AMP will induce it, most surprising is that this blocker of the cyclic AMP pathway will induce it. And so we're continuing other projects from here to try to validate that and look at actual transcription factor binding of stereogenic enzyme genes within the fetal testis, which is a bit of a challenge, and so stay tuned. But it leads to sort of a conundrum, as Dr. Jackson pointed out, that we have a PKA pathway that's super important for stereogenic enzymes, but the PKA pathway is also really important in this primary hedgehog pathway. And so we think it's really interesting to think about sort of isolation of those pathways by the primary cilia, and so we think that perhaps the PKA pathway within the primary cilia is confined there, and perhaps now after listening to Dr. Jackson, we might find some other G protein coupled receptors within the cilia that would be interesting. But with the hedgehog binding, that PKA pathway is blocked, and we think that there perhaps is another ligand that would induce some other GES-coupled ligand or ligand receptor to induce the PKA pathway and help the hedgehog pathway induce these stereogenic enzyme genes. And with that, I just wanted to share acknowledgments of my lab. Ambarasi Kothandapani did a lot of this work along with some previous members, Samantha Lewis and Jessica Newell, and all of the mice that help us along, and a lot of excellent collaborators, Humphrey Yao's lab and Colin Jeffcoat in particular, and of course the funding to do the work. And so with that, I'd be really happy to take any questions and I really appreciate your attention. Thank you. So if anybody has any questions, if you could please go up to the mic. In the meantime, I was just gonna ask a quick question. You were talking at the beginning about the decrease in testosterone levels and the graph that you showed was mostly at older ages. Is there any, is that happening also during the younger years when men are typically more, I'm gonna call them childbearing, but they're not bearing a child, they're... Reproductive age, yeah. Yeah, that's a good way to put it. Thank you. Yeah, that's a good question. So it's not thought that the testosterone levels are much decreased in them, but they have shown evidence of defects in spermatogenesis. And part of the problem they think is goes way back to the fetal testis and the programming that occurs in there that other cells are not functioning as normally. It is also hypothesized that fetal androgens will program those adult leydig cells. And so they seem to function okay to some degree during reproductive age, but as men age, they seem to lose that ability and that's caused a significant issue for men later in life. Thank you. Rodolfo Ray, Buenos Aires, Argentina. Thank you, great talk about leydig cell function. I was wondering if GLE pathway is also present in tertiary cells in the fetal time. My rationale is that, as you mentioned, impaired testosterone production by the fetal testis can be due to a leydig cell problem, but it can also be due to whole testicular problem. And SF1, as from the very early work by Keith Palka, is known to be involved in tertiary cell, also in such a cell differentiation. Yeah, that's an excellent question. So SF1 is definitely super important in both cell types. What's interesting is that at this time when fetal leydig cells develop, SF1 expression increases significantly in those cells. And then in sertoli cells, it stays about the same and maybe even decreases a little bit. The hedgehog pathway is not present in the sertoli cells during fetal development. However, it is thought to come back in the adult testis and also in germ cells. So it seems like even in the adult life, there is an interesting communication between germ cells, sertoli cells, and leydig cells. And we also know that during fetal development, sertoli cells and leydig cells have a lot of other crosstalk, including activins, TGF-beta, other TGF-beta family members, platelet-derived growth factor. So I think that there's a lot more going on than just the hedgehog pathway. Thank you. So I believe we're out of time, but do you have a quick question? I don't know if it's quick. Go ahead. This is Vincent Fong from University of Cincinnati. So I just have a question. We were talking about the cholesterol and how it affects the hedgehog signaling smoothing. And so I do some work on cholesterol metabolism. And so with the upper gation of cholesterol, down gation of cholesterol, you can change hedgehog signaling, I'm sure you know. So then would that have implications in terms of all the metabolic syndrome and metabolic dysfunction that we have if a parent or a mother during gestation had alterations or extra high cholesterol or low cholesterol for some reason? And how would that affect in terms of what you're talking about, development of fetal lytic cells, or even how may that affect the adult testes and testosterone production and spread out just that way? Yeah, if I understand your question, the basic question is, changes in cholesterol levels in the mother and how that might affect the fetus in the end? Is that basically? Yeah, going the other way around, in the cholesterol, changes in levels of cholesterol and how that may affect the lytic cell or the pathways that you were talking about here and the amount of genes that are testosterone production. Okay, so there's a couple ways I can answer that. So we know that defects in cholesterol synthesis, like some kind of genetic mutation, has a profound impact on fetal lytic cell, but obviously it'll have a profound impact on cell division in general, just with the membranes and other things. The fetus itself is, at least the lytic cells are quite protected from maternal issues with cholesterol. They learn very quickly how to make their own cholesterol because they rely on it so much for the rapid cell proliferation that's going on. So those fetal cells develop the means to de novo synthesize their own cholesterol relatively quickly. But any defects that they might have would significantly impact the downstream effect. I hope that answers your question. Thank you very much. Thank you. I'd like to introduce our next speaker, Dr. Christian Weiss. He is a professor at the Department of Medicine and Diabetes Center at University of California, San Francisco. He's also the director at the UC San Francisco Nutrition and Diabetes Research Center. He got his MD and PhD at the University of Paris, Soud, France, and he completed his postdoctoral fellowship at the Rockefeller University in New York. He'll be sharing with us a melancortin 4-receptor signaling in primary neuronal cilia and how that regulates appetite. Please welcome Dr. Weiss. Good morning. Thank you very much for the invitation and thank you to those who are here. So my, I have no financial relationships. So today my presentation will be about MC4R and basically I will tell you about the role of alterations of primary cilia, how alteration of primary cilia function leads to obesity. I will discuss the role of leptin melancortin system and melancortin 4-receptor in control of body weight and show that MC4R localizes and functions at the primary cilia. As we've heard from Peter, primary cilia is a complex organelle which has its own structural proteins, as well, several components, as well as transduction components. And a number of those components have actually been isolated and found due to the fact that mutations cause a disease that affect the function of the primary cilia called ciliopathies. So mutation in certain of those genes, such as R13B or BBS, cause disease. And it turns out that a number of those ciliopathies actually have obesity as one of the phenotype. So alteration in some components of the primary cilia cause obesity. In particular, MOM syndrome due to mutation in ILPP5A cause obesity, mental retardation, retinal dystrophy, and micropenis. Carpenter syndrome due to mutation in RAP23 cause dactyly, polydactyly, mental retardation, and obesity. Probably the most well-known of those disorders is body beetle syndrome. Body beetle syndrome causes retinopathy, polydactyly, polycystic kidney disease, hypogonadism, and obesity. Interestingly, body beetle syndrome can be due to mutation in 22 different genes that actually encode proteins that form the BBO zone complex, which is specialized in transport to and from the sub. So the question is then, with respect to primary cilia and obesity, in which tissue and cells are primary cilia required for long-term control of endothelial stasis? And do ciliopathies lead to obesity by compromising the development or the function of the implicated tissues or both? So to answer this question, the best done in mice, what can be done is remove one of the genes that is specific for the primary cilia, leading to loss of primary cilia in tissues. Now, if you do that germinally, this is embryonic lethal, because primary cilia is absolutely essential for development. But you can also remove, if you have a conditional allele, like IFT88 flocs, you can actually remove it specifically in tissues or inducibly in adults. And so what phenotype do you obtain? Well, early on it's been shown now a while ago that if you actually ablate the primary cilia in an adult mouse, so in every, in the entire adult mouse, the main and major phenotype, and by doing this, for example, by using an inducible ubiquitous tree that can be induced by tamoxifen, in the adult mouse, the major and unique phenotype you will see is severe obesity. After one month, hyperphagia, massive hyperphagia, and obesity. This indicates that the primary cilia are required for energy homeostasis in adult mice. And so this is not a developmental phenotype. Now, more precisely, what tissue is actually implicated? Well, it turns out that if you actually remove the primary cilia exclusively in differentiated neurons, this also causes obesity, basically showing and indicating that expression of one or more receptors at the primary cilium of differentiated neurons is required for long-term regulation of energy homostasis. Now just to, so for my colleagues, neuroscientists who sometimes have never heard of the primary cilia, this is sort of what a primary cilia looks like on a neuron. So what pathway could be actually involved in this system? Well, the major pathway that controls body weight in energy and long-term energy homostasis is actually the leptin-meta-protein pathway. And this pathway is the pathway in the arcuate nucleus of the hypothalamus that responds to the hormone leptin, which is a gauge of adipostitial level. Leptin engages anorexigenic neurons, PAMC neurons, which produce alpha-MSH, and engages AGRP neurons, which are anorexigenic, producing AGRP. Now the importance of this system in body weight regulation is exemplified and demonstrated by the fact that mutations in those components of the system cause severe obesity in humans and in mice. Mutation in leptin cause severe obesity. Mutation in leptin receptor cause severe obesity. Mutation in PAMC cause severe obesity, both in humans and mice. The major integrator of this system is a receptor that is engaged both by alpha-MSH and AGRP, alpha-MSH being an agonist, AGRP being an antagonist. And this receptor is a class A G-protein copper receptor called the aminocholine 4 receptor. This receptor is basically engaged differentially by the two ligands that are controlled differentially by leptin, making it a major integrator of the peripheral signal of obesity, of inequality. Excuse me. MC4R couples to GS to activate unlike cyclase. And this activation reduces food intake. MC4R is expressed in multiple neural population. But it's shown that its expression in the paraventricular nucleus of the hypothalamus, the PBN, is necessary and sufficient for control of food intake. MC4R is absolutely essential for body weight regulation. Loss of MC4R leads to hyperphagia and obesity. Indeed, in humans, mutation in MC4R are the most common cause of monogenic obesity, accounting for about 2.5% of severe obesity in humans. In addition, MC4R is one of the primary hit in GWAS for obesity. In mice, MC4R, non-mice are severely obese. And heterozygous mice have an intermediate phenotype. The major hint of a possible role of MC4R in expression of MC4R to cilia came from specific ablation of the primary cilia in the paraventricular nucleus. Now, in this experiment, the IFT88 Phlox mouse was injected with an adenovirus-expressing tree directly in the PBN. And this leads to massive increase in body weight, the same that you see when you actually ablate the primary cilia from the entire brain, indicating that the PBN is one of the major locations underlying the obesity due to loss of primary cilia. So the primary cilia of hypothermic PBN neurons is required for control of food intake and body weight in adult mice, leading us to question the fact that could MC4R be located and function at the primary cilia. Specifically, is the primary cilia of MC4R neurons required for control of body weight? Does MC4R localize to the primary cilia? And does MC4R function and signal at the primary cilia? To answer the question, is the primary cilia of MC4R neurons required for control of body weight, we basically determine the effect of primary cilia ablation at MC4R neurons. Using a specific tree that we engineered, knocking in tree in that locus, and using it again to remove IFT88 and therefore ablating primary cilia from MC4R neurons, we see both in male and female, and a massive increase in body weight. I have an issue with the pointer. OK, sorry. Pointer's back. So both in males and in females, we observe a massive increase in body weight, an early onset massive increase in body weight. And this increase in body weight is due to increase in fat mass. And this increase in body weight is due to hyperphagia, as we have a massive increase in food intake. This indicates that a primary cilium of MC4R neurons is required for control of food intake and body weight. Now, interestingly, actually, this phenocopies completely the loss of MC4R in mice. So we really have the exact same phenotype when we lose the primary cilia of MC4R neurons than if we lose MC4R in mice. So is this a developmental phenotype? Are the MC4R neurons lost in this model? It turns out that no. We can actually label the neurons, and we find that the number of neurons is conserved. So it doesn't appear to be a developmental phenotype. The neurons are also still activable. When we inject activating DREAD, a receptor that can be activated by a compound such as CNO, we see that we can still activate the neurons and decrease food intake by activating those neurons in the mice that have lost the primary cilia at the MC4R neurons. So this says two things. This says that probably the loss of primary cilia leads to a functional defect rather than a developmental defect in those neurons. So now the next question, of course, is does MC4R actually localize to primary cilia? So initially, we looked at this in cell culture, in ciliated cells such as IMCD3 cells, kidney cells, MEF cells, fibroblast cells, and retinal cells. And every time, and we transfected these cells with MC4R fused to GFP. And every time we did this, we could see that MC4R co-localized with the ciliary marker acetylated tubulin, demonstrating that MC4R could actually be expressed at the cilia, was localized at the cilia in cell culture. So looking at this in mice is more complicated. This is a small hydrophobic GPCR. There's no antibody expressed at very low level. And as for many of those, there's no antibody available. Therefore, we actually created a fusion MC4R-GFP by targeting the MC4R locus and modified it so that we had basically expression under the control of the MC4R promoter in the locus of a fusion protein. When looking at the PVN of these mice on a wide-field microscope and regular magnification, nothing can be seen, which was disappointing. However, going into a much higher magnification and actually looking at confocal microscopy, actually MC4R was detected and co-localized with the unlike cyclase tree, which is a primary cilia marker for neurons. Demonstrated that, in fact, MC4R in vivo is expressed at the primary cilia. Interestingly, this expression of MC4R at the primary cilia does change. In particular, during postnatal development, we first see an increase of expression. And expression is more difficult to detect in adults, although it is still visible. So this does indicate that there might be dynamic changes in the localization of MC4R at the primary cilia. Now, is the primary cilia required for MC4R activity? So the question here is, what is the effect of ligand activation of MC4R following primary cilia ablation? So to test MC4R function in vivo, we need to place a cannula in the lateral ventricle. We let the mice recover. And we fast the mice for 16 hours, and then inject a specific activator of the melocortin 4 receptor, which will decrease food intake. So we then give food back to the mice. And a mouse that is injected with a control will, after this, will eat about 1.2 gram, while a mouse that has been injected with the specific MC4R agonist will actually have a decrease in food intake. So to determine whether primary cilia ablation will affect MC4R activity, we take, again, a mouse that has a flux allele for IFT88. We ablate the IFT88 and, therefore, the primary cilia from the PVN. And then test for the response to MC4R. And we find that ablation of the primary cilia in the PVN ablates the response, the acute response to MC4R in those mice. So the next step was to determine whether analyte cyclic signaling in the primary cilia of MC4R was required for control of body weight. I told you that MC4R signals to GS and to analyte cyclase. For this, to inhibit specifically primary cilia analyte cyclase, we used a receptor called GPR88, which is a cilia-relocalized receptor which couples to GI and, therefore, inhibits analyte cyclase. And we use a constitutive active form of this receptor. So we now have a constitutive active form of a cilia-relocalized GPCR, which would, therefore, inhibit analyte cyclase in the primary cilia. Now, we introduced this receptor into an adenovirus that is tree-dependent. So that, basically, we can now express. When we express this, when we inject this receptor, this virus in a mouse that has an MC4R tree, this receptor will be expressed only in the primary cilia of the MC4R neuron. So we now have a receptor that inhibits analyte cyclase, specifically in the primary cilia of the MC4R neurons. And what you can see here is the following injection, how this receptor specifically localizes to the primary cilia. When we do this, what we see following injection, and then we follow the mice, and we can see that there's a massive increase in body weight. And there's also an increase, massive increase, in food intake. So inhibition of analyte cyclase signaling, specifically in the primary cilia of MC4R neurons, leads to obesity. So here, we also, in addition to that, not only does it lead to obesity, but it also leads to combining now both approaches, where we inject a virus and then test for the response to MC4R. We also find that after ablating, after blocking analyte cyclase, specifically in the primary cilia of those neurons, we also inhibit the activation of MC4R. So in summary, MC4R is localized to the primary cilia. Loss of deletion of primary cilia in those neurons leads to hyperphagia and obesity. And inhibition of ciliary GS signaling in those neurons also leads to hyperphagia and obesity. So the summary and the take-home message is that the primary cilia is essential for the long-term regulation of energy on those stasis. MC4R localizes to the primary cilia. The primary cilia is essential for MC4R function in vivo. And we believe that altered expression of function of MC4R at the primary cilium may be a common path towards genetics predisposition to obesity in humans. So loss of function of MC4R is sufficient to lead to obesity. And a number of mutations that affect the primary cilia lead to obesity. We believe that most of those mutations actually cause obesity by affecting the functional localization of MC4R at the primary cilia. In particular, human obesity candidates that we are thinking of include at least ADCY3 itself, the analyte cyclase of neurons, but also INPP5A, as well as a gene called RPGRIP1-like, which interestingly is located at the FTO locus. For the FTO locus, it's the major GWAS loci for obesity. And we believe that RPGRIP1-like could actually be the gene that is affected by this loci. And this is a major structural component of the transition zone of the primary cilia. So future question, how is MC4R activity at the cilium interpreted by the cell? That is really important. What is it? And how is actually the energy balance encoded by the primary cilia? Are MC4R neuron primary cilia length, content, and responsiveness physiologically regulated? And how do MC4R neurons integrate the long-term ostasis signal conveyed by signaling at primary cilia and the short-term feeding signals conveyed by classical neuron signaling? So how do you integrate those MC4R-dependent chronic changes and those rapid changes that are conveyed by classical neuronal signaling? So one interesting beginning of a story here is that we find, for example, that after calorie restriction, there is an increase of MC4R at the primary cilia. So that is sort of maybe telling us something about possible role of dynamic changes of MC4R localization at the cilia in physiological contexts. Thank you very much for your attention. Thank you. So it's now open for questions. Go ahead. Thank you, Christian. Beautiful talk. I just wanted to ask about actual process of ciliogenesis itself. And do you think actually MC4 receptor signaling might be involved in the formation of the cilia? Because if you lose MC4 receptor, you lose the primary cilia. Is that correct? Did I misunderstood you? No, if you do lose MC4R, you do not lose the primary. So the primary MC4R does not seem to be implicated. So there's no regulation in the actual receptor signaling in the formation of the cilia itself? So no, loss of MC4R does not seem to impact actual ciliation. But regulation of MC4R activity might lead to changes in MC4R localization and changes in cilia length. Thank you. So I actually had a sort of a related question, but vice versa. So when you ablate cilia, does that affect MC4R stability at all? Or I mean, is it still there? Is it still? No, we do not have any indication that it affects MC4R stability or expression. So if you express MC4R in cells that don't have cilia, what happens? It doesn't transduce anything? So OK, so most of the work on MC4R activation has been done in 293 cells. And if you express MC4R in 293 cells, it will go to the membrane sufficiently to be activated by alpha MSH or inhibited by HRP. But do they have cilia? They don't have cilia. Then I don't understand why does it have to be in the cilia to actually transactivate something, to transduce signaling? Well, we think that basically this compartmentalization of signaling allows for compartmentalization of the MC4R function. So that basically signaling from the cilia, and there's a recent cell paper by Jeremy Ryder that shows that compartmentalization of CIP signaling in the cilia leads to different effects than CIP signaling on the cell. So it's coupling with different. So it's called different PKA, and this leads to a specific signaling from the cilia. And I think that's really key, and maybe I should have shown that better. And this is really where we want to go, is what is the signal, and how is it interpreted differently by the cell than signal that would be coming from the membrane. That is essential. And I think I have sort of a related question, because that was sort of something I wondered about, too, is when you knock out the, when you get rid of the cilia, are MC4R and adenylate cyclase 3 still co-localized? Well, there's no cilia. So usually you see them at the cilia. So no, you can't really. Yes, and so the problem is those protein express that low level, and the volume of the cell is so large that you can't really detect them anymore if they are not concentrated at the cilia. OK. And then I also have a question I just wanted to make sure I got to ask. So when you knock out the cilia, of course, you see this hyperphagia. Is there also effects on just lipid handling or adiposity in storage? So say if you knocked out the cilia, and then you took the mice, and you limited how much they could eat, would they still deposit fat and gain weight? So the phenotype is pretty similar to what you observe in mutation of MC4R, right? Yes, so if you per-feed MC4R mice, non-mice, you will see a decrease in the butt. Of course, you will maintain decreased body weight. You still will have a little bit higher body weight, though, due to the decrease in energy expenditure. So I think that will conclude our session. Can we please just have a round of applause for all of our speakers once again? I very much appreciate it.

ciliary FFAR4

adipogenesis

insulin secretion

glucagon secretion

primary cilia

reproductive tract disorders

MC4R signaling

obesity

leptin-melanocortin system

body weight

energy homeostasis

genetic predisposition

food intake