My name is Lee Chan. I'm a pediatric endocrinologist and reader in molecular endocrinology and metabolism based at Queen Mary University of London. My co-chair is Dr. Kalvajit Chowdhury who is an assistant professor of medicine and pharmacology physiology at the University of Rochester Medical Center New York. He unfortunately can't be here in person but will be with us online. We have three great talks ahead of us so without further ado I would like to introduce Dr. Rachel Perry. Dr. Perry is assistant professor of medicine endocrinology and cellular molecular physiology at the Yale University School of Medicine. Her laboratory applies stable isotope tracers to reveal how changes in systemic metabolism can affect tissue specific metabolism and outcomes in cancer, sepsis, obesity, and diabetes. So her talk is entitled Putting the Heat into Obesity and Weight Loss. Thank you. All right well thank you so much for the kind introduction and to the organizers for the opportunity to present here. I'm really delighted to be able to speak to you and thank you so much for being here in one of the last sessions of the day. So I do have a disclosure to lead off. I am one of the patent holders for some material that I'll be discussing today novel 2,4-dinitrophenol formulations and methods using the same and I will be showing some data from one of these formulations in this talk. Here's our QR code and opportunity for questions so feel free to use that questions and feedback. So I'll be discussing meal thermogenesis and work that we've been doing on the mechanisms and effects of meal thermogenesis. So meal thermogenesis has been shown in a number of both preclinical and clinical studies to be primarily induced by carbohydrate and protein rich meals. That refers of course to the increase in body temperature that we see after eating a meal high in carbohydrates and proteins and I think you know a lot of us have had this experience that you eat something high in carbs you may feel a little warmer afterward and this is a this is a thing. In a sedentary individual human or rodent it accounts for about 30% of total daily energy expenditure. Obviously that percentage goes down in an active individual but you can see here in in a resting subject you know this 30% is a pretty good fraction of daily energy expenditure and we know that small differences in energy expenditure can make a big difference in body size and so thinking about you know differences induced by meal thermogenesis of up to 30% could have a big impact on whole body metabolism. And we know based on epidemiologic studies that meal thermogenesis can be blunted in obese individuals and so if an increase in energy expenditure is blunted in those with obesity this would certainly beg the question could defects in meal thermogenesis contribute to the pathogenesis of obesity? And considering my subject and who I am you can imagine the answer to that question. So that's you know one of the things that we wanted to examine in the studies that I'll be sharing with you today. Could this blunting in meal thermogenesis exacerbate obesity? So really the driving question behind the first studies that I'll show you today is what is the mechanism by which refeeding increases body temperature in lean but not obese individuals? So we thought that it might have something to do with leptin and that is because leptin of course is one of the canonical hormones that is altered with both refeeding and obesity. So plasma leptin concentrations this is a study in lean rats and we can see that plasma leptin concentrations drop around 90% in fasted rats that were without food for 48 hours and they bump right back up to fed levels when we refeed the animals let them eat ad-lib for just two hours. And so this seemed to us to be a good candidate to explain some of the thermogenic effect of food. Another candidate of course could be insulin but we thought that that was likely not responsible because of insulin resistance that occurs in obesity. And the data I'll show you certainly are consistent with leptin being responsible at least for some of meal thermogenesis in animals. So to answer that question first we simply infused leptin to see if it had a thermogenic effect. Here you can see we did fast refed studies shown in blue as well as fasting animals infused subcutaneously with leptin to increase leptin concentrations to what was measured in refed animals. And you can see that both refeeding and infusing leptin to what we measure in refed controls increased body temperature by a couple degrees Celsius. So this at least is consistent with a potential effect of leptin to mediate thermogenesis. We really wanted to test that by getting rid of the meal-induced increase in plasma leptin concentrations. And so we studied ooglucknack knockout mice. These are mice defective in a nutrient sensor such that they don't secrete leptin after a meal or on an obesogenic diet. These animals have about a 70% reduction in adipose specific OGT concentration or protein expression. And it was important to us that we locate the expression of the knockout to the adipose tissue because that of course is where leptin is primarily secreted. When we studied these these mice and subjected them to a hyperglycemic hyperinsulinemic clamp. So in these studies we're infusing glucose and insulin rises. We think this is the best approximate to postprandial conditions because you have hyperglycemia and hyperinsulinemia that responds to that hyperglycemia. As shown on the left we see an increase in plasma leptin concentrations only in the wild type animals. And it's absolutely flat in the OGT knockout animals. And so this allows us an opportunity to test the effects of this postprandial leptin in phenotypes that we're interested in. And what we see here is that when we when we study these OGT fat specific knockout animals that do not secrete leptin upon postprandial conditions as shown on the left. They absolutely fail to show a thermogenic effect of food. Body temperature does not rise during a hyperglycemic clamp during under these postprandial conditions. But if we inject leptin to sort of circumvent this defect in postprandial leptin secretion they do show a thermogenic effect. So it's not due to leptin unresponsiveness but due to lack of secretion of leptin under postprandial conditions. So at this point we feel pretty confident that leptin is involved in causing feeding induced increases in body temperature. But what is the mechanism for that? To start to get at that question we infused leptin at several different doses in starved rats. These doses were chosen so as to mimic short-term fasting as compared to prolonged fasting shown in green here. Refeeding shown in white and supraphysiologic concentrations as are seen in obesity shown in yellow. And what we see is a dose dependent increase in plasma catecholamine concentrations primarily epinephrine with these increasing doses of leptin. And so could this epinephrine have anything to do with the thermogenic effects of leptin? We tried to answer that question by either refeeding animals or infusing them with leptin. We do see an increase in plasma epinephrine that is concordant with an increase in body temperature. Both when we refeed the animals and when we infuse with leptin. But if we block leptin's effects using a tunnel all a beta-1 adrenergic antagonist we no longer see any thermogenic effect of either refeeding or infusing with leptin. And so these data suggest that epinephrine is in fact responsible at least for most of that leptin induced increase in body temperature. So at this point we think that leptin promotes adrenomedullary catecholamine secretion and that that is responsible for at least part of the thermogenic effect of food in rodents. But the final question in this first part of the study is how does epinephrine promote meal feeding induced increases in body temperature? We hypothesized because we know that catecholamines act on brown adipose tissue to do a lot of things. And because brown adipose tissue can be thermogenic we hypothesized that these two may be related. And so consistent with this we see an increase in brown adipose tissue long-chain acyl-CoA concentration. So that is consistent with an epinephrine induced effect to stimulate brown adipose tissue lipolysis in these animals. And it's concordant with an increase in body temperature. So to test that possibility we performed intrascapular badectomies. So that removes about two-thirds of the animals endogenous brown adipose tissue. And we see that that reduces epinephrine's effect to increase body temperature again by about two-thirds. This effect seemed to be due to lipolysis because when we treat with systemic aclostatin, which is an inhibitor of lipolysis, it's an inhibitor of ATGL, this aggregates the ability of epinephrine to increase both body temperature and long-chain acyl-CoA. But we can overcome that effect by infusing liposin, intralipid infusion. So if we give back fatty acids, even if we're inhibiting systemic lipolysis, we restore that increase in that long-chain acyl-CoA as well as the thermogenic effect that we see in terms of increasing body temperature. So let's switch gears a little bit and look at some data on why these effects may be blunted in obesity. So here we took lean and obese rats. These are the same rats before and after a high-calorie Western diet to render them obese at about 450 grams. And we can see that body temperature, the thermogenic effect of food, is completely lost in the obese animals. Fasting does not reduce body temperature and it does not increase body temperature further in the refed obese animals. So body temperature starts high and stays high in the obese animals. And that's what we can see here. To test the impact of adrenomedullary hormones on these phenomena, we studied adrenolectomized rats that were either infused with low or high dose replacement corticosterone to take corticosterone sort of out of the equation. And we can see that both low and high corticosterone infused animals had a similar blunting of the thermogenic response to both fasting and refeeding. So corticosterone is not involved in this phenotype. But when lacking catecholamines, namely epinephrine, which we think is responsible here, we again see body temperature starts low and stays low in both lean and obese animals. And so we think that that can be the explanation for what we're seeing here. We have low body temperature in both lean and obese animals that do not have catecholamines. But in sham operated animals, that's where we see the dynamic range. Body temperature starts high and stays high in the fasted sham operated animals because of their hyperleptinemia and hypercatecholamemia, if that's a word. The final conclusion we made from these data is that there's a threshold effect for the impact of leptin on catecholamines and on body temperature. In lean animals, when they fast, they drop below a certain threshold, which we think is around 1 nanogram per mil, 1 to 2 nanograms per mil in terms of plasma leptin, at which point epinephrine drops enough to be responsible for a reduction in body temperature. But in the obese animals, they never get below this threshold. Leptin concentrations do drop a little bit, but not enough to have a significant impact on plasma epinephrine that can explain the changes in body temperature that we see here, or that we don't see in the obese animals. So finally, we ask, can this axis between leptin, catecholamines, and body temperature be restored? So we subjected obese animals to a very low calorie diet, about a quarter of their typical daily caloric intake. And we see that it normalizes plasma leptin concentrations and restores the ability to drop below that threshold during a fast. And that's what we see here. We're getting below that 2 nanograms per mil threshold. This is correlated with similar changes in plasma epinephrine. We see plasma epinephrine really doesn't budge during a fast in obese animals, but we're able to restore that responsiveness to reductions in leptin with a very low calorie diet. And this plays out in a reduction in body temperature in the very low calorie diet fed animals that drop below, that drop their leptin concentrations below that 2 nanograms per mil. And you know, I want to emphasize these these rats aren't back at their high school body weight. They're still overweight by rat standards. But by inducing that amount of weight loss, we're able to lower plasma leptin concentrations below a threshold and allow them to respond to fasting. And the data shown here really speak to that effect. We see above a certain threshold when we compare body temperature to plasma leptin concentrations, the curve is flat. And that's because when you exceed that threshold, that between 1 and 2 nanograms per mil, there's no longer any responsiveness in terms of either plasma epinephrine or body temperature to changes in plasma leptin. And so the summary of part 1 of this talk is that in lean animals, reductions in plasma leptin and increases in refeeding promote adrenomedullary catecholamine secretion, which is responsible for increased bat lipolysis and increased body temperature. Whereas in obese animals, upon a fast, they don't drop below 2 nanograms per mil in plasma leptin, such that adrenomedullary catecholamine secretion doesn't change or doesn't change much, so that lipolysis is unchanged and body temperature does not change under postprandial conditions. Leading us to conclude that meal-stimulated increases in leptin are necessary and sufficient for postprandial increases in body temperature through beta-adrenergic agonism, probably beta-1 from data I didn't have time to show you here. Upon weight normalization, obese rodents regain normal feeding-fasting variations in leptin, catecholamines, and temperature. And both increases in plasma epinephrine and stimulation of adipose tissue lipolysis are required for postprandial increases in body temperature. Which leads us to ask, can we develop therapeutic interventions based on increasing energy expenditure? So considering that, as I mentioned before, meal thermogenesis is blunted in obese subjects. These are just four of the many studies out there. Not important to know the details of the study, but suffice it to say that thermogenesis is blunted in obese subjects. People have explored the possibility of increasing mitochondrial oxidation for the treatment of obesity and insulin resistance. The first dinitrophenol paper, that's an uncoupler, was published in 1933. And follow-up studies have shown that uncoupling with DNP does, in fact, improve insulin sensitivity. So here we can see an improvement in hepatic insulin sensitivity in obese rats treated with DNP. How do uncouplers work? Well, they uncouple the proton gradient, which is required to drive ATP synthase. Because when protons bind to the DNP molecule, the product can simply diffuse back through the mitochondrial matrix without making ATP. And so when this happens, the tissue of interest revs up its rate of mitochondrial oxidation to try to compensate for that energy deficit. Unfortunately, we can't do this very safely. So here you can see a graph of DNP fatalities. They trail off after the 30s when it was taken off the market by the FDA. But with the rise of the internet in the early 2000s, we're back to having some DNP fatalities. And these are on-target effects. You can see maximal temperature in subjects who died of DNP toxicity is, in many cases, over 105 degrees Fahrenheit. And that's because of excessive uncoupling. You can only uncouple so much. So we hypothesized that targeting the effects of DNP to liver would safely and effectively ameliorate obesity-associated hepatic steatosis and insulin resistance. So we first did that by adding a methyl group to DNP. When it goes to liver, the methyl group is cleaved off by the cytochrome P450 system. And so that allows it to be safely handled, such that the toxic to effective dose ratio is improved by 20-fold. And that's the case because of the safer pharmacokinetics of the drug. Peak plasma concentrations, which are related to toxicity, are much lower with DNPME. But it does work. And so this suggested that by manipulating the pharmacokinetics of DNP, we could create a safe and effective oral mitochondrial protonophore. So our second generation, which is what we have the patent for, was controlled-release mitochondrial protonophore, or CRMP, which is DNP with a sustained release coating. That's an oral formulation given to rats in peanut butter. It revs up mitochondrial fat oxidation by about 70%. And it can do this safely without raising body temperature substantially, because it only uncouples in the liver. So we see on the left, DNP raises body temperature in a dose-dependent manner, CRMP does not. But it works. It lowers liver triglyceride concentrations in a dose-dependent manner in obese rats. And so a lot of our future studies were done with this one MIG per keg dose to give us a little range in terms of a dose that we thought would be effective. It too has improved pharmacokinetics as compared to DNP. The toxic to effective dose ratio is 500-fold better with CRMP as compared to DNP because of the lower peak plasma concentration, as well as, and it's hard to see here, but about a three-fold longer half-life. So will it reduce insulin resistance? In fact, it will. We see reductions in fasting plasma glucose and insulin concentrations in obese rats treated with CRMP. This happens without a difference in body weight or food intake. And although, you know, in many cases a weight loss would be attractive, with DNP it could also signal too much uncoupling systemically. And so we were actually happy to see a lack of difference in body weight after treatment. It improves hepatic insulin sensitivity as reflected by lower basal hepatic glucose production, as well as improved insulin-mediated suppression of hepatic glucose production. In addition, it increases skeletal muscle glucose uptake. So it's doing all the right things, improving peripheral insulin sensitivity. And this is because it reduces liver and muscle triglyceride concentrations, as well as diacyl glycerol and novel PKC translocation. PKC epsilon in liver and PKC theta in quadriceps. You might say, why? I said this was a liver-targeted mitochondrial protonophore. Why is it improving things in skeletal muscle? And that's, we think, because of these data. You're doing all the right things by uncoupling the liver. Uncoupling the liver leads to reduced hepatic VLDL export by about 90%. And so that can be, that the reductions in skeletal muscle triglycerides can be likely attributed to this reduction in VLDL export. And these are the last data I'll show you on CRMP in this model. We're most excited, perhaps, about the fact that in addition to reversing hepatic steatosis and insulin resistance, it also seems to reverse established diet-induced NASH. And so we think this could be an attractive therapeutic option both for NAFLD as well as for NASH steatosis. And so to summarize the second part of the talk, mitochondrial uncouplers increase hepatic fatty acid, liver-targeted mitochondrial uncouplers increase hepatic fatty acid oxidation, reducing ectopic lipids, reducing PKC epsilon translocation, improving hepatic insulin sensitivity, and reducing fasting and postprandial gluconeogenesis, as well as reducing VLDL export, which leads to reductions in circulating triglyceride concentrations, reduced intramyocellular lipid content, reduced PKC theta translocation, and improving peripheral insulin sensitivity. And so where do we go from here? This is the last data slide. So the work that we're doing in my new lab is to test whether mitochondrial uncoupling with CRMP has promise for obesity and insulin associated cancers. So we found here in both colon and breast cancer that CRMP slows tumor growth in insulin dependent obesity associated colon and breast cancer in an insulin dependent manner. And so I don't have time to go through this story for you today, but I'd love to talk with anyone who's interested in the work that's being done on uncouplers for cancer. And so the conclusions overall that I hope you walk away with are that leptin is a key mediator of postprandial thermogenesis. Liver targeted and controlled release mitochondrial protonophores are safe and effective methods to increase mitochondrial energy expenditure. Mitochondrial uncouplers are an attractive therapeutic approach for NAFLD, NASH, and potentially obesity associated cancers. And with that I'll close. I'd like to thank my lab, the folks who did a lot of the work here, and I'd like to thank you for your attention and would be pleased to take any questions. Thanks Dr. Perry for presenting so well, you know, quite a lot of data. So this talk is now open for questions. If anyone who has a question, if you can state your name and where you're from and then your question, same with the online people. Excellent talk. I'm a clinician, so I have maybe a bit unfair questions for you, but they are all clinical, okay? One is that I understand that polycystic ovary syndrome, there is also a problem with meal-induced thermogenesis. So can you elaborate maybe of a possible link? Why is that in PCO? So I missed the what syndrome? I apologize. Polycystic ovary syndrome. Yeah, so absolutely. I mean, I think mechanistically, a lot of this could be similar. Unresponsiveness to, inappropriate responsiveness to leptin is certainly a problem in that syndrome. I would be very interested in doing a similar curve to what we showed in folks with PCOS with different body weights to see if that relationship between leptin and epinephrine and body temperature could be similar. It's possible that that curve may just be shifted. But I don't know. I think an important question, though. So the question really was whether androgens would have an effect on this because these patients have higher androgen levels, these females. Certainly possible. Not something we tested in this study, but absolutely possible, and we haven't ruled out. I think one of the challenges with the study that I showed was we initially thought that leptin and that axis that I showed you was responsible and sort of started and ended there. So the second question, also a clinical question. We have sometimes patients, what they call gastroenteric hyperhidrosis. These are patients who eat and they are sweating so much that it's embarrassing for them. They can't eat in public because it's so bad. And we try to give them some inhibitors of the beta-adrenergic pathway, which is not a good treatment because they have lots of side effects. So what could be the cause in these particular patients and would you see any other treatment than this for that? Well, I think this may not be a particularly satisfying answer, but I think a question that I would have that may be unanswerable at this point is is that effect, the hyperhidrosis, due to tiny changes in body temperature, do they get warmer with meals? And that's a question that's hard to answer with our existing technology to measure body temperatures. It may not be sensitive enough to pick up those differences or is the issue that the body responds inappropriately to the same increases in body temperature such that the hydrosis response is greater. And I think that question would probably have to be answered first, likely with implanted temperature probes to really be able to pick up what temperature are the sweat glands seeing and are they responding appropriately? I think it's too early to know if this axis may be involved or not, but it certainly could be. So, okay, so the last question is maybe a bit silly, but your BMI is a spectrum and also people with the same BMI could be medically obese and other ones just very muscular and therefore their BMI may be higher. So I'm thinking that this lack of response to food intake in terms of temperature, could this be like a sign that you are actually obese or not? So when you are on this spectrum, if you don't change your body temperature, that means you should lose weight while if you change, that means you're okay. I think absolutely, I think it's very likely that the body changes its set point. And so we did some unpublished so far, but follow-up studies to really ask that question and say, all right, if we do the very low calorie diet study and we don't put them on a one quarter very low calorie diet, it's maybe a 10% reduction so that eventually they lose weight, but they're still certainly obese. We see that the system has reset its temperature. So at that point, they don't respond to that little bit of weight loss. Whereas if they're only 10% overweight and then they lose 10%, they seem to retain their sort of set point as far as what's an appropriate temperature. And so I think there are a lot of unanswered questions as far as what changes that set point? Because I think it likely does. Thank you. Thank you. So we have quite a few questions online. So great talk, Dr. Perry. Could the effects of leptin or CRMP on thermogenesis be dependent on blood glucose levels? So that's a great question. CRMP probably not because we have performed, at least acutely, hyperinsulinemic euglycemic clamps to clamp plasma glucose concentrations in both treated and untreated animals. And we still see the same improvement in peripheral glucose uptake as well as suppression of hepatic glucose production. That said, it's almost impossible to perform those studies chronically clamping plasma glucose. And so certainly it is possible. I think the reductions in plasma insulin that we see almost undoubtedly are due to slight reductions in fasting and primarily postprandial blood glucose concentrations. The effects of leptin do appear to be due to both glucose and insulin. So if we make animals hyperinsulinemic without making them hyperglycemic, so again, the hyperinsulinemic euglycemic clamp, we see some secretion of leptin, but it's not nearly to the levels that we see under postprandial conditions. And so for the leptin story, I think both hyperglycemia and hyperinsulinemia are necessary to get maximal plasma leptin secretion. So another question is around, so obese patients have, well, I'm not sure this, have higher thermogenesis, would they have higher thermogenesis all the time due to higher catecholamines, or is there a resistance? But I think that we're saying there's a resistance here. We're saying there's a resistant phenotype to increases, but they do, at least in our rats, they do start higher and stay higher. So there is both a little bit higher basal as well as resistance. The other issue is, is a calorie of heat a calorie of heat, where if total body mass is larger, you need more increase in temperature to burn the same fraction of weight. And so there's sort of another level of resistance that comes in in the obese setting. And one last question on the mitochondrial encouplers and its sort of ability to work on NASH with advanced fibrosis. So we have not tested that question. In the study that I showed you, we do see inflammation, we see increased transaminases, we see markers of NASH histologically, but not advanced fibrosis. I'm hopeful that it may have an effect, but we just don't have data to support or refute that. Great. Any other questions from the room? So I think with that, thank you so much for a great talk and thanks everyone for all your questions, both online and here. So next speaker I would like to invite is Dr. Peter McCormick, so who is a reader in molecular pharmacology, also at Queen Mary University of London and a colleague of mine. I've been told to say that he is from Atlanta and this is his first talk in Atlanta after many, many years. And the McCormick lab studies the molecular underpinnings of disease focusing on G-protein coupled receptors in these mechanisms. They use a mix of biophysical, biochemical and cellular models to study GPCR function and Peter's talk today is entitled The Structure of MC4R in Satiety Signaling. Thank you very much, Lee, and thank you very much to the organizers for giving me the opportunity to tell you a little bit about what we do in my lab. I'm not sure the laser pointer's working, but just want to say I have no conflicts of interest or financial disclosures. This, for those that don't know, I'm at Queen Mary University of London. This is where we're located. We have three campuses and I'm along with my colleague, Lee, at this wonderful campus downtown over by St. Paul's. Okay, I'll talk a little bit about a short introduction of G-protein coupled receptors and why they're still interesting to study. I'll talk a little bit about one receptor, melanocortin-4 receptor and its recent structure that helped inform our own functional studies to understand the mechanism. I'll also then tell you, if I have time, at the very end, a little short story about 5AC2C receptor. Yeah, the laser pointer, I wasn't sure if, I guess this was the red one. Okay, cool, thank you very much, great. Yeah, there we go. And then just try to bring it all together at the end. So, for those that don't study these receptors and for those that probably use or study them in physiological settings, just a reminder, these are the largest membrane protein family in mammals. They bind everything from light photons, hormones to proteins and peptides. Typically, they have seven transmembrane domains and they're involved in a variety of biological responses and pathologies. And then this is just kind of a canonical activation mechanism. There are some variations therein, but essentially you have a ligand that binds the receptor. You get activation of the receptor that then causes a conformational change that drives an exchange of GDP for GTP in this heterotrimeric G proteins, alpha, beta, and gamma. And then you've got downstream effectors that drive signaling mechanisms. But for today, we'll focus mainly just on the interaction of ligands, a receptor, and activation of the receptor. So, just to reiterate that these continue to be very important drug targets with 30% of all drugs approved by the FDA acting on G protein-coupled receptor systems. And this has been consistent for the last 20 years that this 20 to 30% of all approvals is around a GPCR. But the beauty is there's a lot of these receptors and there's still plenty of drug targets to be explored within this large superfamily. And my lab is not a structural lab but we're a mechanisms lab. And we use this mantra, this is a quote I took from former head of Genentech many years ago, that the prospect for drug discovery and understanding drug discovery is all about understanding mechanisms. And that's the approach my lab takes in understanding how G protein-coupled receptors work and being able to target them and exploit them for therapeutic purposes. So today, I want to talk a little bit about two receptors involved with obesity. A bit preaching to the choir here, as they say, but this is obviously a chronic pandemic that's been brewing for some time. These are some staggering numbers, two billion adults overweight, 650 million obese. These numbers in children are only worse with the recent COVID pandemic. So it's a serious major problem. My father's a clinician who runs a cohort on the border of Mexico and Texas. And something like 80% of the patients have BMIs off the chart. So it's really unbelievable, staggering numbers that we've got to get a handle on. And my attitude is the following. So this is from a study from now a few years back, but just demonstrating that many people say, oh, it just changed your lifestyle. Well, it's not that simple, right? Obviously, we know there's set points in the body, and this is what just changing lifestyle does exercise alone. This is a change in a very low diet here, and these are pharmacological interventions. So pharmacological interventions are as good as, right, change in diet. So you can imagine some sort of combination therapy might be attractive down the road. So this is the approach my own lab has taken, and I think our own data and the clinical trials out there with some of these compounds, and certainly the new GOP1 combinations look very attractive. So the two receptors that my lab have focused on in this, this is a hypothalamic circuitry of feeding, and I'll first start with telling you about the melanocortin-4 receptor, and then finish with this 5-AC2C receptor that's just upstream. So the melanocortin-4 receptor is commonly known as a master switch of satiety and energy homeostasis in the brain. It's currently a drug target for anti-obesity treatment with a recent drug approval of setmelanotide, which will be one of what I'll focus on today. It's been known for a long time that you have loss of function mutations in the melanocortin-4 receptor that can cause obesity, and gain-of-function mutations which provide relative resistance in these patients who've got low BMI and, in some cases, food disorders. What's interesting about the melanocortin system in general is that you have an endogenous agonist and an endogenous inverse agonist, or antagonist. And what's interesting is with the endogenous agonist, which is alpha-MSH, you drive activation of the receptor, and this provides a sense of telling, simplistically, that the body's full, that you don't wanna have any more food. You're satisfied. Alternatively, if you've got release of the inverse agonist, AGRP, it's gonna bind to the receptor and then tell the body, hey, it's time to feed, it's time to eat, okay? So there's a very important balance, and that'll be an important theme throughout. So recently, just at the beginning of the pandemic, there was a crystal structure of melanocortin-4 receptor by Roger Cohn and Ray Stevens' labs, and they found, and this was with a synthetic antagonist, so SHU-9119, and they found, interestingly, that a calcium molecule seemed to be located within the orthosteric binding site of the receptor, but they didn't really know if this was just a function of the crystal, if it was important for a function of the receptor, and it had been in the, well, sort of concomitantly, I was working with a collaborator at the Weissman Institute, Moran Shelly-Banami, who solved the cryo-EM structure. In this case, she did it with the active receptor still bound to the heterotrimeric G-proteins, and sure enough, she used a set melanotide, the recently approved compound in the clinic for treating obesity, and sure enough, she also found that a calcium molecule located within the orthosteric site, suggesting, again, that, okay, maybe calcium's indeed important for the function of this receptor. In the literature previously, this is a binding study done by the lab of Ego Rinken out in Estonia. They had done a fluorescent binding assay and demonstrated that if you took away divalent ions in general, you could impact ligand binding. Again, this theme of somehow ions were important for function or binding the receptor. My own lab, in conjunction with Moran, went in and tried to understand exactly what the role of calcium was here in terms of functional receptor and activation of receptor. One of the things that we noticed, which was quite unusual, so there's been plenty of ions related with GPCRs over the years, but what is very unusual is that you've got actual direct interaction and chelating of the ligands with the calcium. It's almost like the calcium is a prosthetic group, if you will, within the orthosteric site, which is quite unusual, which again suggested that somehow removing the calcium, you might impact function of the receptor. To test this idea, we measured cyclic AMP accumulation after activation of the receptor, and this is with two different agonists, so this is endogenous alpha MSH and the one in the clinic that I mentioned, cephalanotide, and sure enough, in the absence of calcium, you get a dose response that's in the range of, in this case, around minus eight, and you add calcium, and you get almost a two log shift in the EC50, suggesting that calcium really plays an important role in sensitivity and the ability of the ligand to bind to the receptor and drive activation of the receptor, and this was true both for the endogenous and the synthetic ligand. Interestingly, and I can have you focus on the right side here, which is the endogenous inverse agonist, we saw the opposite, where in this case, calcium, so we're treating with alpha MSH here as the agonist, and what we pre-treat with the inverse agonist, AGRP, and we see in the absence of calcium, we actually get a leftward shift, so AGRP, in the absence of calcium, appears to work better, okay? Now, I'll remind you, for those that don't know, AGRP's quite a large molecule, so clearly the binding mechanism is gonna be much different than a small peptide-type molecule within your esteric site, but the fact that we saw an improvement in the absence of calcium was really quite striking, and I'll come back to that with a little hypothesis at the end. So, one of the other things that the structure revealed was a toggle switch, so for the non-gurus in the lab, in the room, the G-protein coupled receptors, when they undergo activation, have a large movement in transmembrane domain six, and this allows an exchange of GDP to GTP, and a large movement of the G-alpha, okay, underneath. So, you see this movement, this in gray here is the inactive TM6, and this is the TM6 in the active structure, okay? So, you've got a large swinging out of this transmembrane domain, and what we discovered from the structure was, in collaboration with Moran and Masha Niv, also in Israel, was that the ligands were actually touching a leucine here, which then was interacting with the tryptophan on TM6, that are facilitating this outward swing. So, it's a toggle switch, if you will, and this toggle switch has been seen in other G-protein coupled receptors, and just showing you here in cannabinoid one receptor. So, the next thing we wanted to check was, okay, was there any differences in alpha MSH versus in sentinel antide in the ability to recruit G-proteins, or in the ability to recruit arrestin? There had been an association with, demonstrated that arrestin recruitment is very important for this obesity phenotype. So, we just wanted to look and see, what do these two agonists do in conjunction with this? So, that's showing here on the right. So, this is G-protein recruitment in the middle. I lost my pointer. There we go. This is G-protein recruitment, and you see no major differences between the two ligands, and both were able to recruit beta arrestin. This is done using a nano-leuk complementation assay. So, this is just showing you the homology and shared amino acids across the entire melanocortin family, and melanocortin-4 is on top here. One of the things that we discovered from the structure was that this leucine-133, which is that toggle switch I mentioned to you, is conserved across the family, except for one of the members, and that's the melanocortin-1 receptor. And in the literature already, it was known that this SHU-9119 in melanocortin-1 is actually an agonist. So, it turns out that in MC1, this leucine is a methionine, and we think that the methionine, that lack of contact, is now the reason that this normally antagonist becomes an agonist in that receptor. So, that was one of the key features that we learned from this. And then the second was that all of the amino acids interacting with calcium are conserved across the family, which again suggests an important role for calcium in all these receptors. I'll come back to that shortly. So, the next thing we wanna know is, could we test the dependence of calcium on some of these mutants? So, here we made, we essentially mutated all the amino acids that were shown to chelate to calcium. And we first, you can see right away, the idea being, if calcium's important, and you reduce the ability to chelate to the calcium within that prosthetic orthosteric group, are you gonna be able to impact function of the receptor? So, is calcium really essential? So, without removing all three, let's do one at a time. Do we happen to see an impact on the ability to produce cyclic AMP? And sure enough, you can see here that for D122N, and it didn't matter on the agonist, you could severely restrict cyclic AMP production. That was true also for this D126Y, and I'll come back to both those mutations shortly. And then finally, this E100A, which is the third site of chelation for the calcium, you can see, again, a very dramatic decrease in the activity of the receptor. Now, I'll also come back shortly between this difference between alpha MSH and setmelanotide, which at the time, we didn't understand ourselves, but subsequent studies helped clarify that. So, this is just showing you the calcium dependence itself experiment on the D122N mutation, and you can see very nicely that you still maintain some calcium dependency because you still presumably got the other two sites for chelation. So you haven't completely gotten rid of calcium. It is still there, but something about that coordination and lodging of the receptor makes it now a partial agonist. And then the E100A, again we saw this difference between the two ligands, and I'll touch on that shortly. A subsequent study helped inform us why we see a difference between alpha MSH and set melanotide for this particular mutation, at least what we think. So importantly for those 122N and 126Y mutations, those are actually found in patients. Those are clinical mutations. So one of the things that was important was that for at least for one of those, they responded with set melanotide. So a demonstration, a good example of personalized medicine, that it's obvious that those patients with that particular mutation can be treated with set melanotide and should be able to respond. Because indeed they have a severe obesity and an increase in BMI. And in addition, it supports the idea that this calcium prosthetic group is important for function of the receptor. So what's our working hypothesis on this calcium connection with these two ligands for this receptor? What we think is that in the presence of calcium, local concentration of calcium, right, you would favor an alpha MSH binding. And lower concentrations of local extracellular calcium might favor AGRP binding when it's present. So we think it might serve as an allosteric mediator between these two endogenous ligands at a local concentration level. So after our study came out, two subsequent studies came out with structures. The first from Brian Kobielka and Patrick Shearer on melanocortin 4-receptor, essentially repeating our work along with adding two other ligands. And they saw very similar things. They saw the calcium located for both agonists and they also saw the calcium dependency in cyclic AMP that we saw. So it's always nice to have your work validated by other folks. And in addition, Eric Zhu out of Beijing did very extensive structures of melanocortin 1-receptor. And also they saw calcium dependency for melanocortin 1-receptor. Again, supporting this idea of a general theme of commonality across the family of receptors. And if I go back to that difference we saw with the E100A between the two ligands. So this is from the Brian Kobielka and Patrick Shearer paper. And what you can just appreciate is the lines connecting calcium to the ligand here in green and the receptor here on the right side are different with this particular ligand versus this is sep-melanotide on the right and alpha-MSH on the left, NDP-alpha-MSH. And you can appreciate. So what we think is that there are more connections between the alpha-MSH versus the sep-melanotide. And that might explain the differences between the results that we got with that E100A and with the calcium dependency between the two ligands. That's our working hypothesis and that was one of the things they pointed out in their discussion as well. So that was melanocortin-4 receptor. But I want to just finish with just upstream of melanocortin-4 receptor is the 5AC2C receptor which then releases and activates PalmC neurons which are responsible for that alpha-MSH secretion. So I just want to tell you a little bit about what we've done with 5AC2C over the years. So 5AC2C, well GPCRs in general, one of the challenges of them is that orthosteric sites are common across the family. So for example in the melanocortin family that orthosteric site with the calcium is relatively shared across the receptor. So specifically targeting one receptor in a family at the orthosteric site is more of a challenge. So this is just an example for a 5AC2 family. They have very similar orthosteric binding sites and obviously you're going to have a lot of side effects depending on what indication you're after. And it's very challenging to get a very specific molecule. Now this is a molecule, urcacerin, which it was in the clinic and quite effective anti-obesity molecule. But it was an agonist if 5AC2C had some mild side effects and was pulled for a small population of people that were developing cancer. We got interested in this idea of perhaps targeting a different part of the receptor might reduce those side effects and reduce any concerns of safety. So the beauty of allosteric sites, and I'm showing you our particular allosteric site, sits above the orthosteric site, is that they are unique in a given family. So they allow an opportunity to specifically target one member of a family. So with this approach we first identified in a high-throughput screen a molecule that we published a few years ago with a group out of Madrid. And we could find in rats it was very effective at weight loss, both with rats limited access to food and with continuous access to food that uses just in the presence of our compound. But the the screen was more of a classic throw a million or so compounds at the receptor and see what comes out. So there was no structural information. It was very blind and relatively lucky if I may say. But the affinity of the molecule was a little bit low. So more recently a group out of Galveston, a Catherine Cunningham group, identified another allosteric modulator with an improved affinity and they've been showing it's very effective for treating smoking cessation and drug addiction. So we got motivated to go back and revisit our allosteric compound using, in addition, there were five new structures that came out for 58C2 receptor. So basically went back and underwent a structural-based drug design concept or a program to try and identify improved allosteric compounds against 58C2C. And this is, I'm just showing two of those, demonstrating that and sure enough we can see very nice allosteric effects with these two compounds. So this is an increase in calcium release in the presence of the allosteric modulator and this is with serotonin, the endogenous agonist, and this is with another molecule, more modest increase. But interestingly we also looked at G-protein recruitment and we have slightly different profiles for the two compounds. So we're now taking these forward in animals in collaboration with Li-Chan here to see how effective they might be compared to our previous compounds. So hopefully I've convinced you that calcium ions are important cofactors for melanocortin family and melanocortin 4 receptor in particular and they're essential for agonist activation and that local calcium concentrations might serve as an endogenous mediator between the ability of alpha MSH to activate the receptor and AGRP to act as an inverse agonist. And finally our data support that set melanotide can be given in these patients with the D122N and we're pursuing others. We've screened quite a number now of patient mutations to demonstrate where it could be effective and where it won't most likely won't be. And finally that structural based design of allosteric compounds, particularly allosterics targeting 5AC2C receptor can be an effective alternative to direct activation for anti-obesity drugs in the future. I just want to say thanks to all my collaborators, in particular Dr. Moran Shaila Benami who I work closely with on the structure of melanocortin 4 receptor. Li-Chan who was part of that study and gave us invaluable input. Dr. Leslie Howell for making the compounds for 5AC2C and then the two graduate students, very talented, working on both projects, Vinita Trinidad and Amandeep Gill. And I'm happy to take any questions. Thank you very much. Thanks Peter. A great walkthrough of what we are current understanding of the structure of MC4. Any questions from the room? Whilst we wait for questions from the online. Hi. I'm Eileen Hanyalalu, Imperial College London. Thank you Peter. It was a beautiful talk. I wanted to ask about the calcium cofactor in MC4. Do you think it might be possible to pharmacologically target that site, even potentially even like as a by topic ligand in just to, you know, improve this melanotype responses, especially with the loss of function mutants and the inpatients? Thank you very much. That's a great question. So we were working with Masha to try to model, she's trying to model that same idea. Could we put in a small molecule which either won't need that or will mimic that space or that charge to try to provide either improved affinity and better specificity between, and particularly we're targeting some patient mutations which were null. So can we create an agonist which will now make them either partial agonist, get some response out of those receptors. Now obviously for some of the mutants it's a trafficking defect and Michel Bouvier's got some great chaperone small molecules for those patients, but I think there's others where there's situation where just what you said would be a great approach. So we're hopeful. Thank you. So this is a question from our co-chair. Could changes in energy balance like lean versus obese phenotype affect neuronal calcium levels and bias the binding of MC4R with its ligands? Great question. I mean that I would love to be able to test that and I would love to be able to test changes in calcium locally like that. That's one of the things if anyone has any ideas I would be delighted to see because that model is still that a model in these testing. Another question online is great talk. How can you ensure that enough calcium concentration is present to maintain activation of MC4R? In the literature the extracellular calcium is in the right range. If I recall it's between 0.5 to 1.5 or 2.5 millimolar reported. So those are the ranges that we use in our experiments. My guess is it is local concentrations do fluctuate and I think this is part of part of the interest to understand. Any more questions? If not thank you so much for a great talk. Thank you. So our next speaker who I'll invite up is Dr. Alessia Perino who is a senior scientist in the Laboratory of Metabolic Signaling in Lausanne, Switzerland. Her research is on the role of the bio acid response GPCR TGR5 in metabolism and to aiming to unravel the normal functions of bio acids signaling in metabolic tissues in order to develop novel strategies to target obesity and type 2 diabetes. Her talk today is entitled Role of Bio Acids in Controlling Satiety. Thank you very much for the kind introduction and it's a great pleasure for me to have the opportunity to present our data at this exciting meeting that is almost over. So I have nothing to disclose and yeah the QR code last one for this session. So as yeah one of the main research topic in our lab are bio acids. As you might know bio acids are cholesterol catabolites. They are produced in the liver starting from cholesterol through a complex series of reactions that is just briefly summarized here and these reactions they give rise to primary bio acids and these primary bio acids in hepatocytes can be conjugated to amino acids specifically taurine in mice and taurine and glycine in humans. So these bio acids are then stored in the gallbladder and they are released in the intestine upon food intake and they were originally discovered as lipid solubilizers. So they were important for the absorption of lipids but then we know that these primary bio acids in the intestine can also be let's say they are substrate for the gut microbiome and they are converted into secondary bio acids and only 5% of these bio acids is excreted in the feces. The rest is taken up and recycled back to the liver and these bio acids can be reused in the so-called enteropathic circulation. We know that bio acids they are not only present in this enteropathic cycle but they can also be secreted and found in the systemic circulation and this was discovered about 15 years ago. This is a study in humans but then this was yeah also present in mice and we know that in humans bio acids are among the most abundant plasma metabolites in response to a glucose challenge. So when we eat and when we treat mice and humans with glucose we have a rapid and substantial increase of bio acids in the systemic circulation. So this means that bio acids are not only lipid solubilizers but are actually hormones that can signal the presence of nutrients to all the organs of our body and this is feasible because bio acids can act and signal to different receptors that can be either GPCR which I don't need to introduce anymore or nuclear receptor. So TGR5 and FXR are the most studied receptor for bio acids and our lab is mainly focusing on studying the role of TGR5 signaling and we know that TGR5 is mainly activated by secondary bio acids and binding of secondary bio acid to this GPCR triggers a PKA cyclic and P signaling response that mediates the non-genomic effects of bio acids and on the other hand primary bio acid mainly bind to FXR to nuclear receptor to trigger the genomic effects mediated by bio acids. So these two different receptors they are expressed on different cells of our body and the activation or balance and expression of these two different receptor can modulate metabolism in general and this is just a summary of all the different let's say metabolic responses that can be regulated by bio acids and by the selective activation of these two different receptors in both mice and humans. So we know that bio acids in the hypothalamus can reduce food intake through TGR5 and this is actually the topic of my talk today and we know that bio acids they can also activate increased energy expenditure and reduce body weight, they can increase insulin resistance, they can modulate glucose and lipid levels in the liver and in the intestine. Through FXR bio acid can also regulate their own level so they can modulate bio acid overload. We know that in the gut bio acid can increase gut hormone secretion especially the incretins and they can also increase gut regeneration especially in the context of colitis. Bio acid can also increase aging and lipolysis in the white adipose tissue and they control immunity and they reduce inflammation. So as I mentioned before the main the topic of the talk today is going to be how bio acids they can escape the enteropathic circulation if they can reach the brain and if so if they can have a role there. And to answer this question we basically used black six mice so wild type mice and we study the presence of bio acids in the brain and especially in the hypothalamus after physiological feeding. So we as I mentioned before bio acid they can escape the enteropathic circulation and they can increase in the systemic circulation upon feeding. So our question was can this bio acid reach the brain and if so can they have a function there. And so we measured bio acids in hypothalamus as well as in the brain but these are the hypothalamic data and we see that in fasting condition the bio acids in hypothalamus are low but when we refeed the mice in a physiological condition for 30 or 60 minutes we see a rapid accumulation of bio acids in the hypothalamus and the different colors here they correspond to different bio acid species that can be either primary, secondary, conjugated or unconjugated bio acids. And bio acids levels are still up in the hypothalamus after 60 minutes after feeding. So this means that yeah we hypothesize that bio acids can have an important role there. And we of course thought about food intake while because bio acids they can they are released upon food intake and they we thought that it was kind of a possible feed negative feedback regulation of feeding and so we use the bio acid mix which is just a mix of different of these bio acids and we administer this bio acid mix to wild type mice and we monitor the food intake over a period of 24 hours and we noticed that administration of this bio acids mix orally could significantly reduce the food intake in these mice. So this I also don't have to introduce I will just briefly mention that yeah as we know the hypothalamus is the most important one of the most important regulators of food intake and in the hypothalamus especially in the armpit nucleus we find these two different classes of neurons that are called HRP and PY and POMC. So the HRP and PY are orexigenic neurons and their activation increases food intake while the POMC neurons they are anorexigenic neurons and their activation blocks food intake. And this is mainly thanks to the activation of second order neurons and MC4R. So we know that bio acid they can reach the hypothalamus and we then wanted to know if bio acid had an important role in this particular hypothalamic region so the armpit nucleus and we first started to yeah we thought we hypothesized that this function could be mediated by TGR5 and we started by analyzing the TGR5 expression in this hypothalamic region and we used two different techniques so the brain punches and the RNA scope. And the brain punches technique we just sliced the brain and we punched the armpit nucleus and we tested the TGR5 mRNA expression in this region and we used the TGR5 wild type mice so control mice in white and as a negative control full body knockout mice for this receptor and we could observe that TGR5 is actually expressed in this particular region and the levels do not change with the feeding condition. We then use also RNA scope which is a kind of fish like technique so we detected the mRNA of TGR5 and again in wild type mice we could get a signal so meaning that TGR5 is expressed in this region and the knockout mice this signal was gone. We then wonder in which cells was expressed TGR5 and to do so we used a reporter mouse model which we called TGR5GFP. In this knocking mouse model, the mouse TGR5 was substituted with the construct encoding for TGR5, F2A, and GFP. So then we sliced the brain and we used the control mice or our reporter mice, and we stained the brains for GFP. So we could find GFP expression in our reporter mice and was not there in the control mice. And then we used different markers and we could find co-expression of GFP and new N. So meaning that TGR5 was actually expressed in neurons in this region. So we then went on by phenotyping our mice. We started with the TGR5 wild type and TGR5 full body knockout mice. And we measured food intake over a period of 24 hours. So we saw that basically when we deleted TGR5 in the whole body, we could get a significant increase in the food intake response. And this was true both for the night phase and the day phase. I mentioned that TGR5 is present in neurons, so we then removed TGR5 in neurons to see if this feeding response was neuronal. And to do so we used what we call the TGR5 synapsin-free mice. So we bred our TGR5 flux mice with synapsin-free mice to delete TGR5 in all the neurons. And we could, with this mouse model, we could recapitulate the phenotype that we observed in the knockout mice. So removal of TGR5 in neurons, in the whole body neurons, significantly increased the feeding response. So meaning that neuronal TGR5 was important to mediate this response. So we then wondered which were the neurons that were driving this response. And we started by analyzing the mRNA expression of the AGRP and PY and POMC neuropeptides in both fasting and feeding condition. In both models, so our TGR5 knockout model or the synapsin-free model, when we measured the POMC mRNA expression, we couldn't see any difference between the control mice and the mice lacking TGR5, either in the whole body or in the neurons only. But this was different when we looked at AGRP and an NPY expression. So in the wild type controls, feeding induced a significant decrease in AGRP and NPY expression. And this decrease was blunted when we lacked TGR5. So basically in both the TGR5 knockout models and in the synapsin-free model, these mice, they always had high levels of AGRP and NPY and they could not modulate this response by feeding. So we then used also pharmacological approach and we used what we called INT777. So this molecule is a bilacid analog and it activates selectively TGR5. So we used again wild type mice, black six mice, and we administered this bilacid analog both orally or intracellular ventricularly. And then we again analyzed the expression of POMC or AGRP-NPY peptides. So again, in this case, the TGR5 activation did not impact on the POMC mRNA expression. But again, we found an association between TGR5 signaling and AGRP-NPY. So in this case, both oral administration or intracellular ventricular administration of this bilacid analog could, let's say, reduce the expression of both AGRP and NPY. So then, yeah, these data suggested that TGR5 was important in regulating food intake thanks to the heat section on AGRP neurons. But to really prove this, we generated other transgenic models. So in this case, we crossed our TGR5-PLOX mice with either AGRP-CRE mice or POMC-CRE mice to delete selectively the TGR5 either on AGRP or POMC neurons. And while the deletion of TGR5 in POMC neurons didn't affect the food intake response, we noticed that the deletion of TGR5 in AGRP neurons significantly increased the feeding. And so this is just the mean. Then we also administered the TGR5 agonist, so the INT777, to both wild-type mice and TGR5-AGRP knockout mice. And while we saw a significant decrease of food intake here in blue, when we administered this selective TGR5 agonist to wild-type mice, this response was blunted in the TGR5-AGRP knockout mice, meaning that TGR5 was really important in these neurons to modulate the food intake response. So as I mentioned before, with our in vivo data, let's say, demonstrated that TGR5 could modulate the transcription of AGRP. But we know that AGRP and NPY, these neuropeptides, they are also secreted. So we next asked if bile acids through TGR5 could modulate neuropeptides, and in this case, AGRP and NPY secretion. To answer this question, we used an in vitro model, the mHypoEN41 cells, and this is an hypothalamic cell line which we used to kind of mimic the feeding and fasting response. And so we starved these cells for four hours to mimic fasting, and then we stimulated the cells for short time points, so five, 10, or 15 minutes with either vehicle or with our TGR5 agonist. And then we measured AGRP secreted in the medium. So in the cells treated with vehicle, the AGRP levels were kind of stable, were kind of high in the medium. But we noticed that when we stimulated the cells with the TGR5 agonist, so when TGR5 is active, we could basically blunt the secretion of AGRP. And what was interesting was that this response was kind of transient, so the secretion of AGRP was again normal after 15 minutes. And we also proved that, we wanted to prove that this was TGR5 dependent, and to do so, we used the shRNAs, either controls or shRNA for TGR5. And so we demonstrated that our TGR5 agonist could block the AGRP secretion in the SH control cells, and this was not present anymore when TGR5 was deleted. So we know that AGRP and NPY, they are secreted, they are present in the so-called dense core vesicles. So these dense core vesicles, they are normally present in the cytosol of the neurons, and they can be secreted. And this mechanism is dependent on the actin cytoskeleton. So actin can form a physical barrier that blocks, that prevents the secretion of these AGRP vesicles, or can be depolarized and can allow the secretion of these vesicles and the AGRP and PY. So then we wonder if this TGR5 dependent blockage of AGRP and PY secretion could be, let's say, mediated through this mechanism. And we know that actin polymerization is regulated by different mechanism, but the RhoA-Roc signaling cascade is one of the most important pathways that can regulate this response. And, yeah, GPCRs can activate this pathway. So we reason that this TGR5 dependent mechanism could be dependent on the RhoA pathway. So we then used, again, our hemipocells, and we measured, we used the same protocol I described before, and we measured the AGRP secretion. And we first, yes, stimulated the cells with INT, with the TGR5 selective agonist, and we can get a decrease of AGRP secretion at the short term. We then used Tiazovivin. Tiazovivin blocks this signaling pathway, so it's a rock inhibitor, and it blocks the actin polymerization. And the co-incubation of INT777 with Tiazovivin rescued this effect, meaning that the actin polymerization is then important for the secretion of this AGRP that is mediated by TGR5. We then also imaged this, and we used the same cell lines, and as well as an NPY, mCherry, plasmid. We transfected the hemipocells with this peptide, and we could see that with the stimulation of the cells, with either our selective TGR5 agonist, or with a bilacid mix, we could have a significant accumulation of NPY, mCherry in the cells, but this is also quantified here. So to conclude, we basically demonstrated that feeding can increase, can, let's say, increase the bilacid concentration in the blood, and these bilacids can actually reach the hypothalamus, and in particular, the AGRP neurons. In these AGRP neurons, these bilacids can activate TGR5, and TGR5 acts through a double mechanism to control the AGRP levels. So on one hand, it can block the AGRP and NPY transcription, this is in the long term, but we also discovered a novel mechanism that regulates the short term, so really the secretion of the AGRP in these neurons. And why we think this is important, it's because we think that bilacid and TGR5 signaling in these AGRP neurons can actually, let's say, be a negative feedback loop that regulates physiological feeling. And with this, I would like to end by thanking my supervisor, Professor Christina Skunjans, all the people in our group at EPFL that participated to this work, the EPFL platforms, and our collaborators who helped either with, who provided us with this selective TGR5 agonist, or with the experience in the neuron part. And of course, I would like to thank you, all of you, for your attention, and I'm glad to get any questions. Thank you. Thank you, Alessia. Very exciting work, and a lovely demonstration of the network and how it feeds back. Thank you. Can I start with the first question, actually? And it's about the bilacid mix. So, is that mix going to be different in terms of obese and lean individuals, or, and does, will that have an effect on receptor signaling? Yeah, yeah, this is a great question. I mean, we know that the bilacid pool changes with obesity, and we also collaborated, so this was, let's say, a back-to-back study. This was on the physiology, but we also collaborated with Daniela Cota in Bordeaux, France, and they studied more the role of these same axes in obesity. And we know that, I mean, what we think, but of course, we need to work more on this, is that, yeah, the bilacid composition in obesity is different, so this can impact on the activation of TGR5, but we also think that in obesity, we have a change of the, let's say, the ability of bilacid to reach the brain. And this, I mean, we still don't know exactly how, but if we compare lean and obese mice, in obese mice, we have less bilacids that can reach the hypothalamus. So we think that we kind of lose this, let's say, feedback control, and this would contribute to the increased food intake, and yeah, obesity in general. But yeah, we need to work more on this. Thank you. Thank you. I just wanted to ask that, not just obesity, but maybe patients who had cholecystectomies, they probably release these bilacids not with a meal when you contract the gallbladder, but like more generally, or, you know, so I don't know the physiology or the mechanism of that, but I'm sure that this would have an effect on these functions, so have you thought what would happen to patients who have cholecystectomy and what would change this regulation? Yeah, I mean, this is a great question. We do not have data on this, but what we think is that, yeah, either the permeability or the ability of bilacids to reach the brain can change, because of course we know that, I mean, bilacid, they can be anti-inflammatory, but when there are too much, I mean, if we have too many, too much bilacids, they are soap, right? So they can also increase inflammation. So it's possible that in cholestatic patients, we actually have an increased inflammation in the brain, and so this could impact, could prevent the bilacids to enter the brain and to activate the signaling that I just showed. Another option is also that, I mean, high levels of bilacid could also desensitize the receptor. These are, we have, let's say, they're conflicting data on this, but it's possible, I mean, TGR5 is a GPCR, so it's possible that if we have bilacid, but the signal is not active, then we don't have the function. So both possibilities are there. Beautiful talk, Juan del Ringo on Cleveland. Following that question, we studied some patients that were undergoing bariatric surgery who had a liquid meal, and as soon as, even before the 30 minutes that you saw in your mice, the bilacid levels were dramatically different in the circulation. Some of those patients had cholecystectomy. So do you have an idea why or how these bilacids can be released to the circulation so fast? Hmm, yeah, it's a good question. Well, the reason why we use the 30 minutes time point in mice is because we wanted to have, let's say, a refeeding, a real refeeding, and the mice, I mean, they eat, but we cannot say, you eat now. You need to give a bit of time to them. What we noticed is that we also try to, let's say, use a test meal in mice, but we noticed that the response was not the same as with feeding. So we hypothesized that, yeah, maybe a neuronal effect from the intestine, I mean, from the gut to the brain could, let's say, potentially be involved, but we don't, yeah, we don't know which one and how it works. Yeah, and following on that, too, so the idea that bilacids are gonna be released and they're gonna reach the distal intestine where most of the, well, these receptors for TGR5 are gonna be present, but that's a long circuit. I mean, that's a, so it looks like something will happen through a circulation directly. Yeah, it could be through, I mean, neurons from the periphery that go directly to the brain, and it could be like a double response. And one more question. Yeah. You showed that in the mouse with deletion of the TGR5 in the AGRP neurons, they have a decrease in food intake. Were there any differences in body weight or like in other more chronic parameters? This is also a great question. So we monitored body weight and energy expenditure because, yeah, our first, I mean, we know that bilacids, they, in a diet-induced obesity, they increase energy expenditure and they decrease body weight, so we hypothesized that, yeah, this could be, let's say, present also here and could be regulated through the hypothalamic, through the hypothalamus, but actually, in physiology, we didn't see any difference in body weight. And we saw a difference in body weight, but in diet-induced obesity, I mean, high-fat diet-fed mice. So we think that, yeah, this difference is probably because this circuit that we demonstrated is really physiological one, so, of course, you don't want to decrease too much your body weight. So it's really like to stop eating, but then, yeah, you need to have, let's say, a high-fat diet or an obesity situation to have an impact also on the body weight. Thank you. Hi, hello, I'm Alessandra Mancini from Boston. I had, so first of all, actually, excellent talk. Beautiful data, so well done. So the hypothalamus, and especially the arcuate nucleus, as I'm sure you know very well, is also a hub for reproduction and pubertal timing. And so, first of all, a question. Were your mice pre- or post-pubertal? Post. Post, yeah. So any thoughts on, like, TGR5 in neuronal plasticity, as you showed about the role on acting? So just a thought. Yeah, yeah, yeah, well, it's an excellent point. We are not really brain experts. This is for this reason we collaborated with, let's say, other groups. We know that bile acids, in general, can regulate male fertility. This is really, I mean, on the spermatogenesis. But of course, yeah, we thought about the HPA axis, and yeah, and I mean, we don't have data, but it makes sense. Thank you. I have one last question from my co-chair. So are circulating insulin and glucose levels changed in the TGR5 knockout mice? If yes, they might impact feeding behavior in these mice. Yeah, yeah, good point. So we checked these, and in child diet fed mice, in the model that we used, we didn't see any difference. We measured glucose, insulin, GLP-1, because we also know that TGR5 activation can increase GLP-1 levels. But these are all things that we know in high-fed diet. I mean, this is really the first study, the first, I mean, the physiological study of bile acids, because, yeah, we know that bile acids and the deletion of TGR5 can increase, I mean, deletion can increase glucose levels and can bring to insulin resistance, but this is only in high-fed diet. Great, thank you so much. Thank you. So with that, I would like to thank all three speakers for some excellent talks, and for all of you for staying right till the end, and just before lunch, so sort of perfectly sort of right that this talk was at the end. And thank you very much for making it to the end of Endo 2022. Thank you.

melanocortin-4 receptor

MC4R

structure

calcium

binding site

receptor's function

sensitivity

affinity

agonist ligands

toggle switch

receptor activation

satiety signaling