 Ingram for today's Steenbock Lectureship. I want to start off by thanking all of the generous donors who contribute to the Steenbock Lectureship. Without them, this lecture series would not be possible. So I invited Holly as my first Steenbock talk for a number of reasons. So because of some AV difficulties, I'm not loading my slides, I'll save them for her wonderful talk tomorrow. But I just want to point out, Harry Steenbock was well known for a lot of his work, not just in vitamin A and vitamin D, but also in the ability to transition how an organism's environment impacts their physiology and also impacts disease manifestation. And I think this work particularly aligns with Holly's research, both her lectures today, which focus on and fossil lipids as a sensor for hepatic lipid metabolism and gut brain access, and also for her work tomorrow, which will focus on nuclear receptors and the regulation of osteoclast and osteoblast formation. And so beyond that though, I think what's wonderful about Harry Steenbock is he was really a visionary in the field. And I mean, there's so many phenomenal researchers out there, but a researcher like Harry Steenbock who had the vision to use his patent money to set up Worf, who I mean truthfully his vision inspired and helped build the careers of faculty like me through the Harry and Evelyn Steenbock fund. And I think that's really one other reason why I invited Holly here. So besides her really phenomenal work in an animal physiology and regulation of disease, she's also been important in having a vision for changing science. And part of that has been done through her work with the UCSF Aracta program for which she was awarded the Martin Luther King Jr. Award through UCSF. And it's really a testimonial to the fact that there are so many phenomenal researchers that Holly is really changing the demographic of science. And so please help me in welcoming Holly to Madison. This is her first time and we are baptizing her by ice. Well, thank you. A couple of things. When I got the invitation and I looked who else had given these lectures, I was telling my husband I better do a good job. And I also got off the plane yesterday and it was it was quite warm here. So my first time in Madison is really been pleasant. Thank you. I've enjoyed all of the talks that before our conversations has been great. I can see that you guys have an amazing metabolism effort here going and it's really enviable. Okay, so today I'm going to talk about work that really started when I went to UCSF and I came out of Jeff Rosenfeld's lab of a lab of about 55 people and had to sort of pick up and decide what to do. And I picked up the textbook of endocrinology and decided to work on a problem based on that. I don't know if I would suggest that for you younger people but it's really what got me into the field of nuclear receptors which I had vowed never to work on in Jeff's lab because there were always these there was a lot of activity and a lot of competing men in that field. And so but I I'm going to tell you about this work it's it's sort of the end of 20 or 25 years and I'm sort of moving in a new direction but I think we've really learned a lot about this class of nuclear receptors. Okay so the nuclear receptors that we have really focused on are these NR5A and NR5A2 their their own little subfamily and they are somewhat unique in the terms of the nuclear receptor in that they have a very large hinge domain and they bind as monomers to a half site and they're always in the nucleus. So one of the things about these receptors is that the pharma basically chose to ignore these receptors because they are important in development and they wanted to figure out how to drug other receptors many of the steroid receptors which you already know about. One of the things that we did with Gary Hammer back in 1999 as well as other people has showed that these receptors are modified by post translational modifications in this large disordered hinge domain and so we but before I go into that data I'm just going to give you a glimpse of where these receptors are expressed during development and they're named also stereogenic factor one because Keith Parker cloned this and thought that it was important and it is important for stereogenesis and liver receptor homolog one which is really not a great name David Moore coined that because it was initially found in the liver. Okay so in development you can see that NR5A1 or SF1 is expressed really highly in the adrenal gland in the gonads and in the spleen and then in the hypothalamus and the pituitary so it really marks the whole endocrine system which is why I was pleased to start working on this as it because of its role in regulating anti-malarion hormone and so it's very prominent in these tissues development and it persists in the adult and then in NR5A2 or LRH1 is found in a lot of the epithelium of the intestine the pancreas and the liver as well as some other places so knocking this out is an embryonic lethal because it also regulates oct foreign stem cells this receptor is does not is not embryonic lethal but does lead to changes in sex development which is how I got into it. Okay and as I said one of the reasons that I started getting interested in this receptor is because of this sexually dimorphic expression with its expression high in males and low in females and in fact this is really important because it's this differential in expression that helps regulate a key protein that's going to cause destruction of the female reproductive tract in males so we worked on that for several years looking at the biology of that but then we get you know one of the things about a nuclear receptor is you want to know how it's regulated and so we sort of approached that two ways we looked tried to figure out okay what is its ligand what is binding in its binding pocket and what about these post translational modifications do they regulate receptor activity this is just to show you the if you make a reporter a knock-in reporter for the SF1 you see it lighting up attesties an adrenal and the ventral medial hypothalamus which is going to be a lot of my discussion for tomorrow so I like to show this right around Halloween or if you've gone to see that play in New York a wicked gets all about green so I just I love this these green the green adrenal is great okay and as I said these receptors are heavily modified by phosphorylation as well as Sumylation and in terms of a substrate for Sumylation which I'm not going to go into the detail I hope that everyone sort of understands about Sumylation Sumylation is a it's a protein that gets conjugated it's a small protein that gets conjugated on to another protein and there's a usually a cycle of it going on and off and what's important is that it's this hinge domain that gets heavily Sumylated and if you recreate this in in vitro assay by adding the really you only need one the complex one complex to get this going an ATP as well as UCP 9 you get this really nice Sumylated pattern and then that of course disappears if you add the protease in one thing that these receptors are is in in vivo and in cells you can easily see that they're Sumylated so unlike other people that work on Sumylation we really did not have to work very hard to show that these were fully Sumylated and here you can just see that if you take the various wild type you can see that they're heavily Sumylated if you make mutations you you get rid of this Sumylation in all of these cell lines and importantly in primary hepatocytes so this this these proteins are heavily Sumylated and now we we pursued this somewhat I wanted to pursue this more but the reagents for Sumylation you can easily do a loss of function you can't do a gain of function like you can with phosphorylation what we do know about the Sumylation of these proteins is that once they get Sumylated you get gene repression and if you get rid of Sumylation all of a sudden you see these gene changes in these what we call Sumo sensitive targets and that's best illustrated here where we were the first to actually go in and do a knock in of the Sumo mutants in SF1 several years ago and we you can see here we in in this mutant we've lost that Sumylation in these embryonic or postnatal tissues here and the the upshot of that was that we would turn on these genes here's sonic hedgehog being turned on ectopically and inappropriately in a testes and the the consequence of this is that we had an expansion of latex cells that produced steroids so we had very very high testosterone in very small testes and we also saw phenotypes in the adrenals so this was exciting in at the time because it was really one of the first in vivo demonstration that this post-translational modification was doing something in vivo and so that's sort of where we've left that story we we wanted a way to figure out how you could regulate Sumylation by pharmacology and we did screens to look at for inhibitors of in our 5a Sumylation and found one but it's very difficult to get a substrate specific inhibitor let alone a substrate specific antibody to of Sumylation so this is these are one of the challenges in the field of Sumylation but so we we know that post-translational modifications are important for the regulation of these receptors and we then at the same time we're asking okay what regulates these receptors in terms of a ligand and to do that we so remember that these receptors were thought to be constitutively active they're in the nucleus and the question is will they be regulated by any ligand and in order to do that I sort of dove into crystallization and went back to my biochemistry roots from graduate school and we then crystallized both the ligand binding domains of SF1 and LRH1 and discovered in well we initially did in 2003 we crystallized the ligand binding domain of the mouse LRH1 and there was nothing in the pocket so I said okay great these are receptors that don't need a ligand but then we crystallized the mouse and the human and the human LRH1 and SF1 and what we found is bacterial phospholipids right in that pocket with the lipid tails just a scooting up into that pocket with a really pretty large pocket for a nuclear receptor and so we found phosphatidylcholine, phosphatidylglycerol, phosphatidylethanolamine all stuck in the pockets and then the question was really are those lipids important they could just be that they come along for the ride during the crystallization and they help stabilize the receptor and at the time when we presented this work at the nuclear receptor meetings we were met with lots of skepticism because this doesn't look like any of the steroids or some of the the the the other ligands for nuclear receptors that had been identified and so we had to keep working on this to convince people that no they really could be bound by a nuclear receptor could be bound by a phospholipid and so and the other thing that's going to become important is that there were human mutations in SF1 that lead to gonadal ovarian impairment and adrenal dysfunction in this disordered loop here that we could never see by structure and that will become important later on. So Ray Blind came to the lab as a as an or well not originally as an arachnid fellow become became an arachnid fellow and he's now an assistant professor at Vanderbilt and it's really carried on much of this work and what he I said we have this problem we have a phospholipid in this pocket what could be the ligand the real ligand and it didn't make sense to me that it would be something as boring as a bacterial phospholipid so he said about and went through a series of phospholipids to look at the phosphonositols and found using a gel displacement assay with really a beautiful assay where you could see the difference in the mobility of the ligand binding domain and you can see here that PIP 2 and PIP 3 are the highest affinity ligand for this for both SF1 and LRH1 for the ligand binding domain and so he you know one would think that if you could find the best drug that would go in and actually alter the activity of these receptors it would be a PIP 2 or a PIP 3 memetic which we still don't have today so this is just showing you the what this is the ligand binding domain of LRH1 so this is a of course alpha helical bundle with this phospholipid the lipid tails stuck up into that pocket and then the PIP 3 the head group sticking out in the mouth of the pocket and we got beautiful electron density of that PIP 3 so we know that it's in there very well ordered and if you now take all of our combined structures as well as other structures done by several other groups including Eric Ortland's group we can see that this is the cloud of the ligand binding domain pocket here this cavity this large cavity and here's the phospholipid tails sitting up there with the various phospholipids in the head groups so we have basically a nuclear protein running around with a phospholipid in it with different potential for maybe modification and what I will say is that in working with Chris Kuchenbacher and and Flederich's lab as well as other labs we we actually tried to figure out okay what happens if you have this lipid in the pocket does it change your ability to interact with co-activators because that that was the standard test that was done by pharma to find all of these synthetic ligands if you ever work with nuclear receptors most of them were all found by screening for recruitment of a co-activator peptide and what you can see here is that even though we have a an affinity difference in terms of peptide recruitment between pip three pip two and here is the APO with with nothing in it you can see that that's a very small differential so in fact the screening that was done by pharma to look for in a ligand in terms of a co-activator recruitment assay really did not work so well and this is this is probably why because we know this is a very high affinity ligand so there's something fundamentally different between these this receptor and other nuclear receptors in terms of that assay what we did find with when we actually got the structure of pip three bound to SF one is all of a sudden we saw that these this loop two three became ordered and that is it precisely where these human mutations are so we really felt very good that having the right ligand in there orders that whole pocket and and where you can now see that mouth of the pocket so we have these lipids in there we've got pip two pip three and one of the next questions was okay are there pip twos and pip threes in the nucleus does that even make sense and work from maybe two decades ago suggested that would be the case and Richard Anderson that was who I was talking to this morning was one of the first to suggest that the whole phosphoanacetide kinase and phosphatase cycle is cool is present in the nucleus and active and I think this might even be from one of this nuclear phosphoanacetide cycle is there this is showing that the pip kinase is present in the nucleus and we now know that that almost all of the kinase is in the phosphatases exist within the nucleus what we don't really know is how how those fossil lipids get loaded into this nuclear protein and I would only suggest that there were two papers out this last year suggesting that there can be lipid droplets that then come out into the nucleus and you have these luminol or nucleoplasmic lipid droplets that are there that can be modified by different enzymes and this is actually just showing one of these droplets in a in a real hepatocyte so there were two different papers so it could be in fact that the that the way that these fossil lipids the original fossil lipids get loaded is through the the protein coming up and grabbing a fossil lipid and taking it out because we do know that if you give these proteins liposome with pip 2 and pip 3 they just go in and grab that right away but we also ray worked for several years to look to ask whether because we know there are kinases and phosphatases in the nucleus could you actually go in and modify the lipid when it is bound to a protein very similar so if you think of these proteins they're almost like there there's like a plasma membrane where you're delivering the head group and it's able to be modified by different enzymes and so that's sort of what I've drawn here here's SF1 with a pip 2 here and what Ray found is that the IPMK which we know is a nuclear kinase it normally of course is regulating an acetal phosphate multi-kinase so it's regulating these these small an acetal phosphates but IPMK works very well at phosphorylating this pip 2 to basically create pip 3 once it's bound to SF1 and we we showed you know it's an it's a correlation but there's a changes in transcriptional gene targets when that occurs and then he showed that p10 can then take this phosphate out to make a pip 2 so one of the one of the ideas that he's that he's really pursuing now is that I an IPMK and SF1 as well as LRH1 form a very nice interact they have very nice interaction and the question is is whether there is changes to recruitment of of other proteins that are really using this head group to get information from so is there a code with these lipids once they're bound into these transcription factors so I in talking to Richard I mean there's going to be other proteins in the nucleus that are bound by lipids it's a really hard problem to to address because we would like to know you know how how do these lipids then really affect transcription okay so Ray is really pursuing that down at Vanderbilt and I knew people come into the lab and when new people come into the lab they have ideas and so I generally say okay let's let's pursue some of those so in fact Diego Miranda came into the lab and his passion was the liver which I knew something about but I mean he really knew a lot about it and you can see wait let's see he is now at Gilead is one of the lead scientists in the Nash program so he came in and said I want to work in the liver and I said okay LRH1 is in the liver and I want to create some some scaffolds where we could perhaps test compounds and and understand a little bit better why how these proteins are connected to fossil lipid metabolism to sort of understand this connection and so he came to me with this that basically from GWAS studies this GWAS studies as well as several others that LRH1 is repressed in patients with fatty liver or Nash so this suggests that you need the full dosage of LRH1 to be around to help offset the Nash coming on so that was great I said this this will work great but there's one problem and that is that basically when we did the as I said when we did the original mouth structure there was a salt bridge right here and no fossil lipid in the pocket and then in the human what you can see is that basically this whole region becomes ordered in part by the phosphate delivered from the fossil lipid to really help stabilize that mouth of the pocket so we have a problem I I want to know about this and I want to know about why fossil lipids are important but Diego wants to study this so we have a problem so we have to overcome that problem and that was one of the first problems that we wanted to to overcome which is to humanize the mouse and figure out how we can study human LRH1 in the mouse and we also wanted to tag these in our 5a's because the antibodies really aren't great for a lot of different assays and then one thing that I thought was really important was to be able to do a gain of function study so everyone you know just we do the loss of function no problem but we can't go back and do a gain of function and I thought this was really crucial because if people are saying oh fossil lipid ligands no way our ligands even important for these receptors I wanted to test a mutant that we had developed from the structure that stabilizes the protein but precludes any fossil lipid from coating in so I wanted to be able to test those side by side with the wild type so and then of course one of the ideas is could we create platforms and identify new targets that would be activated by human LRH1 so we turn to this method and really Diego was the one who said maybe we should try this method so Maury Bernbaum at University of Pennsylvania really pioneered this method where they're using basically you can deliver into a mouse an AAV8 TBG which will get it into the hepatocyte and then you can deliver various you can deliver a protein or you can deliver a cre-recombinase and the nice thing about this whole method is we could flag we can tag our protein and so we could go in and simultaneously deliver cre to knock out the mouse and then re-express the human or re-express a variant and I have this this is perfect I really like this because we're going to use a lot less mice we're going to spend a lot less money and basically once you do this you have a stable expression that that goes out I mean we've gone out to four months and we still have very stable expression using this method so and the TBG gets you to the hepatocyte so this is just the data you can see here if you take this is a GFP if you put in GFP versus a Cree you knock out the mouse LRH1 and you can see these this is a really good Western for LRH1 with these antibodies so they're not really great antibodies but then if you deliver a Cree and human LRH1 you add back the flag tagged human LH1 is seen here and you've eliminated the mouse so this this really works well so the first thing we did is we actually compared it how does taking something out in the adult work versus taking something out earlier so a lot of people in the end that work on liver use the albumin Cree a lot of those people never think about development I do so actually the albumin Cree comes on at embryonic day 14 and that's really early in development and so I I always worry what are we affecting in terms of what compensation is occurring because you're taking it out early in development and and I in this is part why I wanted to do the AV TBG system because I worried about what would happen if you take it out so early in development and in fact so the original knockouts were done many years ago by cleaver's group and and David Moore's group and they really reported very sort of on the the phenotypes were really not very remarkable because they thought they would see a change in bile acids but in fact it was very modest and in fact so here's what it looks like this is on even on a high-fat diet for six weeks you really don't see much of anything this is just a section of a liver so if you now do the same thing where you take it out at six weeks of age and you treat these mice or give them six weeks of high-fat diet you can now start seeing this steatosis in these macro vesicular lipid droplets so clearly taking it out acutely in the adult has an effect and what we start to see is fibrosis as shown here by the cirrus red and some of the other markers so that was that was great we we're starting to see a phenotype when we now maybe we can get somewhere in terms of how LRH1 is connected to phospholipids but probably the most marked thing we see is that even after two weeks if you take you do this manipulation you knock out LRH1 and then isolate hepatocytes two weeks later you see this marked increase in lipid droplets here okay and this just shows you the distribution of lipid these are very large lipid droplets that accumulate so we worked with Mark Hellerstein and and others as well as our mouse metabolic core to figure out what might be the why are we getting these large lipid droplets so we're getting increased lipid storage here we do get decreased fatty acid oxidation so the getting rid of mobilizing these lipids is attenuated but when we looked at other changes like free fatty acid import into the liver triglyceride export is not attenuated and with Mark Hellerstein we did de novo lipogenesis and that is not different so there's clearly a problem we have increased storage decreased fatty acid oxidation and so of course we turned to trying to figure out what more might be going on and we did profiling on these livers both with high fat diet and with chow these data I'm just showing you are from chow because we wanted to sort of mitigate all of the things that occur one a mouse is on high fat diet and what came out of this what came out of this profiling or you get increased genes that are going to be important for lipid droplets but we started seeing changes in genes and elongates and desaturates that are important for polyunsaturated fatty acid synthesis in this I think this is just showing that that what we see on the microarray is real so we decided to do lipidomics with David Silver who Diego had done his undergrad or his graduate work with and Martin Winky at Duke Singapore and we looked at these genes a little bit more carefully and showed that the desaturates the elongates is we're all significantly down both on standard diet and high fat diet so these of course are going to take dietary linoleic acid and one of the important things they do is create these polyunsaturated fatty acid arachidonic acid so and and I highlight that because when we did the lipidomic analysis what we see in the when when we knock out LRH one is we see a loss of all of these it's it's quite impaired the the amount of arachidonic acid derivatives but this is I mean I think I've talked to some of you who do lipidomics and I I really have an appreciation for how complex this is after after just trying to make sense of all the reams of data that we got from these guys from Singapore and this is just shows you show you right here you see this drop in the arachidonic acids fossil lipid species so this is was cool because we at the same time Peter Taunton knows and others had suggested that having the ability to have this diversity about with fossil lipids is important for fossil for lipid storage and so what we would suggest is that without these diverse fossil lipids you you've got decreased membrane fluidity and increased triglyceride accumulation and so that's sort of where we what we think is going on with the LRH one knockout and it connects to fossil lipid we still don't know what the what the key fossil lipid ligand is but at least now we have a function of LRH one with fossil lipid metabolism which I was quite happy with so without LRH one we have a decrease in fossil lipid diversity we get liver damage and steatosis and this just shows you that we can now rescue this with the human LRH one in fact it rescues even a little bit it does very well at rescuing and really quite nicely the pocket mutant which cannot take up a fossil lipid does not rescue so we've got a we've got a connection to fossil lipid and more importantly for me we have a reason for why you would have this large pocket and why you would want to take up a lipid fossil lipid as a ligand okay so at the same time that we were finishing doing this story I was approached by Jim there who is a pediatric GI doc who treats kids with inflammatory bowel disease and one of the things that he wanted to do was his goal really is to come up with new therapies for treating inflammatory bowel disease for kids rather than immune therapy because once you initiate immune therapy it's for life and he felt that rather than going to that immediately because he sees these very young kids with inflammatory bowel disease he really wants to think of new ways to treat inflammatory bowel disease that does not require that you modify the immune system forever so he had worked with a fear client on and learned and really was well very well versed in organoids and these intestinal organoids that as as somebody said today they're not quite differentiated but they're really quite good and and they're they're much easier than hepatocytes because you can culture them and you can culture them for a year or more and I could even culture them I took care of them on Christmas for him and so this I didn't kill them and I didn't contaminate them so I thought these were really cool little things and Hans Cleaver I know has given the talk here and this is really an amazing contribution I think for the field so Jim was we can take mouse organoids and we can create human organoids so he can go in and take out punctures and create human organoids from his patients from healthy and disease patients so one of the ideas that he wanted to look at LRH one and using a villain Cree that's inducible by tamoxifen he can then create these organoids that completely lack LRH one so you can do this in a mouse but you can also do this in an organoid which is quite nice and the other thing is is that we could genetically manipulate these organoids to create an organoid where we could knock out the mouse and then replace it with the human with this human flexileal so that's that's also very nice and then we can take human organoids and what he showed is that you can use an AAV trick to infect with you can do this in mice in the mouse organoid or the human organoids you can then infect that organoid with an AAV system to express now with organoids of course you have a limited lifespan it's not like hepatocytes so every seven days things are gonna go into the lumen they they they change over so it's it's your window of time is much more narrow than it is with the liver and a hepatocyte nonetheless the first thing I asked him to do was to go back and sort of do some work I'm showing that LRH one was in the organoid because Johan Arwick's group had shown early on that maybe it was limited to Crips and they had suggested that LRH one would be important for the gut as well but what we found actually is that LRH one is actually throughout the entire organoid it's it's a little bit more enriched in the crib so I'm using the the the villain Cree and the precursor of tamoxifen the pro drug you can now completely eliminate LRH one in these organoids as shown here and when he did that and he treated the organoids with an inflammatory insult like TNF alpha you can see this is a wild type organoid this is where we've knocked out LRH one and you now see this crypt cell death very prominently so LRH one is a very important for maintaining the integrity of these orinoids and especially for that crypt region he also showed quite nicely that when you knock out LRH one you lose the barrier function so this is a pseudo lumen of these organoids remember it's inside out for those of you who don't know but you can do a vital die and show that without LRH one that whole barrier falls apart as seen here and so now what we wanted to do was do what we had done in the liver which is to replace it with a human LRH one and so that's just shown here where he's replaced it with human using this flexed allele we can also do this with a viral delivery and what you see here is here's a wild type we've knocked it out you see this cell death this is in a mouse intestine so this is not in an organoid and then if you give back human LRH one not even at very high doses you completely protect against cell death and this is just showing in organoids what I'm going to show you is that giving human LRH one is very protective against cell death but the pocket mutant isn't so once again it means that that lipid ligand is important for activity something that I think for me after studying these receptors for as long as I have just knowing that was was really great so we got together with Alex Wang and David Moore who were doing experiments in mouse using this T cell transfer model so this is a model where you you're going in and you're basically inducing colitis so for those of you who know or don't know a lot of people in the field use a chemical induced colitis model it's really not the greatest this this is actually a much better model for inflammatory bowel disease or colitis and because it's it's not chemical and what you see here is that this is the normal mouse without in it's it's not perfect but it it survives all right if we start knocking if LRH one is knocked out you really have disease that that goes up and then this can be restored if you give just even a small amount of LRH one human LRH one can restore improved disease indices and restore survival as shown here should show yeah this is just the weight curves and you can see restoring with the human is here this is without LRH one and here's the fox fox allele so this is great the human works perfectly fine in the mouse and in fact it works a little bit better and so what Jim did is he captured some diseased and healthy organoids from his patients and we asked okay in a human organoid human intestinal organoid they're a little bit more complex to culture because you have you can put them in differentiation media or not but we were really pleased to see that LRH one is throughout this hole in the epithelium of these human organoids and he then insulted these with TNF alpha and asked okay what happens when you give LRH one and you can see both in a healthy and a Crohn's disease setting you can improve viability so clearly human LRH one is is great to have on board or you want your full dosage of LRH one around in the urine testing so what we basically have shown here is that you you improve resiliency when you have human LRH one on board are you over express it slightly we don't know about agonists we have tried all the available agonists we just really don't have a great agonist yet to work in vivo unfortunately work they'll work in cells but not in vivo but what we have found are some unique human targets of LRH one CTRB which now we think if for going forward for a pharma that wants to develop drugs we've got some great targets for them to look at because I think before people were really using mouse targets and these are definitely specific to the human protein okay and then the other thing that we that we started looking at is what are the cell types look like in these organoids or mice that we have taken LRH one out of and not to get overly complicated if Hans Cleaver here this he would have single cell data and as gazillion clusters of all the different cell types within the intestine I'm making this very simple and showing you an intestinal stem cell with an absorptive cell an EEC cell which are these are these peptide releasing cells as well as intracromophen cells and goblet and panaceous cells and there's this you know very stereotypic differentiation pattern that occurs in both the gut and in organoids so what we Jim found is that notch signaling goes down when we get removed LRH one and goblet and panaceous cells go up which is what we might have expected and this is just the data showing an increase in these goblet cells here but unlike some of the earlier work that's been done on notch we had a bit of a surprise in that the EEC cell types go down so this is just and yeah you would have Jim is really patient I don't think I would have had the patience to do this so this is you can imagine this is an intracromophen cell so this is about one out of a hundred cells in the gut epithelium in actually in organoids as well and so I don't know if you can see and there's just these red cells scattered throughout here and he went through and counted in all the different parts of the intestine and what you see is a really pretty marked reduction in the cell type in when we get rid of LRH one and that's just quantified here so this is the normal levels of intracromophen cells and the drop when we get rid of LRH one and we see a drop in these other immature EEC cells so this losing LRH one not only affects causes cell death it causes a breach in the barrier but it also reapportions this the cell types within the gut epithelium and we were excited to see this drop in EC cells because at the same time that we were looking at that in a project full disclosure that I was that that actually was doing with my husband's lab David Julius we we started looking at these EC cells a little bit more specifically and it was really Jim and Nick who got together and used the organoids and used the intestine to look at these specialized gut epithelium cells and look at their sensitivity so these are these are the cell cells that are going to make 90% of your serotonin in your body and remember I just showed you pictures of how few there are in the gut so one out of a hundred is about all you need and you're producing 90% of your serotonin and they're really important because they they as we showed in the cell paper they are really pivotal for that gut nerve interaction and that was one of the questions that they set out to ask is are these really receiving signals and then sending out signals to to nerves and Nick who's just a supreme electrophysiologist worked out this whole system worked out what the receptors are to irritants like to trip a1 and bacterial metabolites to then cause signaling to then cause release of norepinephrine or serotonin onto nerves and it's that signaling that we think is really important and why you develop sort of a gut ache and what Jim really nicely showed is that these EC cells which we've lost a lot of when we get rid of LRH1 are nestled right up against those nerves and and form these synapse synaptic like connections and the real you know the the electrophysiology is nice but I like to go back to whole animal physiology and was stewed briarly we learned how to do the visceral motor reflex response so I know this is I never thought I'd be showing a slide like this starting out from structure crystal structures to this but this is actually you you put a little balloon up the anus of a mouse and then you start stretching it and they actually do this in in for patients because for children that have problems with controlling their defecation or pooping they go in actually and do this assay to make sure that the reflexes are in place this is a this is a innate reflex and so you're basically going in and distending the balloon and then looking at the reflex with electrodes to see how much the the mouse is responding and this give you an idea just as a control if you take one of these bacterial metabolites which is isovallurate and you look at this response you can see here that in fact here's isovallurate really activates this response here and this is the this is in a wild time mice and so here we have our LRH1 knockout and you can see the profound effect of losing 60% of those EC cells has a really profound effect on this reflex so we we really think that you know the the bottom line here is that we I think of all the places that I would be thinking about for drugging the NR5A's it would be in this context which is for inflammatory bowel disease or irritable bowel disease which is to try to increase LRH1 dosage or activity in the epithelium and I don't think that by doing so you're going to have an effect on proliferation there's controversy about that but I we have me seen nothing that suggests that over activating this receptor is going to cause proliferation or cancer so we think that LRH1 is incredibly important for preserving that the epithelial health and in a way if you think about inflammatory bowel disease one of the things it's an immune disease but it happens because there's a breach in the epithelial layer so perhaps drugs to LRH1 would help offset any inflammatory insult so we of course would love to know what the real lipid is in you know what is the real true lipid ligand in these different tissues I don't think we really know that and can a high affinity efficacious drug be identified for these receptors and then just in two minutes I'm going to tell you what we're what we're moving into and we've just gotten funded for which is my favorite thing thinking about sex differences if you come to my talk tomorrow you're going to hear a lot about that but women suffer far more about three to four times more with IBS than men do and so one of the things that we've now gone in and done is use this this VMR to ask what happens if you lose all of your estrogen versus giving estrogen back and you can see that estrogen is really important for this reflex we don't know where estrogen is I mean this is stuff that we're all starting to tackle and we're going to tackle it in a way that we've done with other tissues which is to use some of these tricks so I've meandered from structure well development structure physiology probably sticking to physiology although I love by tomorrow you'll hear my talk where we're going to have to go back to biochemistry so as a scientist I and I think it's you get to do exactly what you want to do and so as you're just limited by what questions you ask so I just I love all of these different techniques that I've used throughout my career so I need to thank the people Jim Diego Myra and Andreas have really worked on the current projects Ray blind and me Suzawa were just instrumental for some of that early work that we did on fossil lipids we have lots and lots of collaborators at UCSF and Baylor as well as abroad and at UC Berkeley that have all really helped contribute to making these stories the best they can be and I cross this I always show this because I cross this bridge to get to work and I it's always the bridges in science that really make it fun and enjoyable and enlightening in my opinion and if anybody is interested this is the first time I've ever worked with my husband we just got a you a one grant to study that got nerve system and visceral pain and sex differences so I'm looking for postdocs so if anybody is interested in this project please email me or find me thank you thank you so much for a wonderful talk we will open it to the floor for questions and while we're waiting I will ask one so you've done such a phenomenal job of characterizing these nuclear receptors in the different tissues they're expressed with their various functions I'm curious you observed in our five in our five A2 in the pancreas I'm curious what it's functionally doing there so I mean people have looked at that and I think what has what Matias Hebrock has actually looked at that and they did a loss of function and I think they were originally thinking that they would see cancer but they don't basically they see a lack of differentiation of other cell types so it had I don't know that they followed that up so closely and I don't know if other people are doing it but I know that it probably has a very important function in the pancreas and potentially maybe they're not using the right models if it's knockout in early differentiation in the hepatocytes so that's probably part of the problem in terms of the phenotype other questions so I think there's evidence that hepatocyte nuclear factor for alpha binds to fatty acid and I'm wondering how big a family do you think there are of nuclear receptors that bind to lipids or lipid precursors and how diverse is that family and other interactions that P par one of P par delta I think our P par alpha has been suggested to bind to phosphatidylcholine and I think I don't know if they had the structure but they had pretty compelling data to suggest that was the case but it hasn't really been it hasn't been followed up it they haven't done as much work as we have to to prove that but they had fairly compelling evidence and I so yes could other receptors be bound by phospholipids probably could other nuclear proteins be bound by phospholipids definitely it doesn't really so that's that's why we were a bit confused because it doesn't directly at regulate those genes it seems to regulate well it regulates these genes that are involved in polyunsaturated fatty acid synthesis not so much fatty acid oxidation it regulates these other genes that have been implicated in lipid biosynthesis but haven't really been worked on so a few but not the standard ones I'd have to look I'd have to look go ahead Mitchell that's a great question so we have not looked and I think that would be something I think that Jim is going to be following up to look at that because you you know because the liver and the epithelium share so much I think it would be great to look at that we just haven't done it yet all right I'm sensing an inpatience for wine which is just outside so let's give Holly one more thanks thank you for a phenomenal talk