Good afternoon everybody. I want to welcome you to PCOS and Metabolic Consequences the Skinny on Adipocytes. I'm Nanette Santoro. I'm professor and chair of OBGYN at the University of Colorado School of Medicine. I'm Rachna Shah. I'm a pediatric endocrinologist and the director of the polycystic ovarian syndrome clinic at the Children's Hospital of Philadelphia. So it's a pleasure to introduce as our first speaker Ricardo Aziz who's going to talk about adipocyte insulin resistance in PCOS relationship with glucose for expression whole body glucose disposal and micro RNAs. Dr. Aziz is an educator scientist executive with over 25 years of leadership experience in higher education research and academic health. He serves as the chief science and strategy officer and executive vice president for operations strategy and scientific affairs at the Lundquist Institute for Biomedical Innovation at Harbor UCLA Medical Center. He served as chief officer of academic health and hospital affairs of SUNY system administration as founding president of Georgia Regents University founding CEO of the Georgia Regents Health System and president of Georgia Health Sciences University. He's also served as deputy director the clinical and translational sciences Institute and assistant dean for clinical and translational sciences at UCLA director of the Center for androgen-related disorders at Cedars-Sinai and founder and executive director senior executive director of the androgen excess and PCO Society. Among other national and international advisory capacities he served on multiple NIH committees chaired the US FDA Advisory Board on reproductive health drugs and served on the oversight committee for the California Institute for Regenerative Medicine. He has received among many recognitions the 2000 President's Achievement Award for the Society of Gynecologic Investigation, elected member in the Association of American Physicians and received the 2014 alumni fellow award of the Penn State University Alumni Association. He serves on the faculties of the School of Public Health, the University of Albany, State University of New York and the Schools of Medicine and Public Health at the University of Alabama at Birmingham. And with that as an introduction we look forward to your talk Dr. Aziz. Thank you very much that was a very much appreciate the introduction. Hopefully I've been told that I'm gonna get pulled out of here at 25 minutes and I'm already it's already ticking and so we'll try to go through this relatively quickly. I'm not going to go into an introduction of what polycystic ovary syndrome is but most of you know that polycystic ovary syndrome is a single most common endocrine disorder of reproductive age women ranging in in prevalence somewhere between 7 and 12 percent using relatively strict criteria broader if you use a broader criteria. This is a studies from a laboratory and to be fair this has been replicated multiple times by other investigators but when you look at adipose tissue and we've chosen to study at a subcutaneous adipose tissue and no it's not just because it's easy to study but because it's actually the largest fat mass in the body and defects in the and subcutaneous adipose tissue will impact the remaining insulin action. We speak a lot about visceral fat but visceral fat in general is a relatively smaller percentage of total body fat and so what you find when you look at adipose sites and I don't know if this pointer is working or not maybe it is maybe not so when you look at adipose tissue and you look at in IMGU or insulin mediating glucose uptake you will find that it is significantly deficient in polystic ovary syndrome patients regardless of whether they're actually obese or not and these actually in this particular slide are non-obese polystic ovary syndrome patient and then when you look at GLUT4, GLUT4 being the principal mediator for glucose transport in adipose tissue you will find that it is deficient in polystic ovary syndrome patients both whether they are have measurable insulin resistance or not relative to matched controls. Controls with insulin resistance also tend to have a lower GLUT4 expression in adipose tissue as you would expect. We've also looked at GLUT1 just as a control. GLUT1 which is not the principal mediator for insulin mediated glucose uptake is in fact not altered in polystic ovary syndrome patients so it appears to be fairly GLUT4 specific defect. There's a large number of studies out there looking at this adipose tissue dysfunction in polystic ovary syndrome impaired adipocyte lipolysis, perturbed capacity of subcutaneous adipose tissue to safely store fat, altered adipogenesis, etc. So I'm not going to cover the gamete of all the different defects that have occurred or been identified in adipose tissue in polystic ovary syndrome. Suffice to say that adipose tissue is dysfunctional in polystic ovary syndrome. So we began in a systematic way to try to study what the defect was and we began really in a simplistic way to look at insulin signaling PI3AKT kinase pathway and we again were trying to emulate some of the studies that had been found in other tissues and this is a study we published a few years ago in diabetes and we looked at the variety of pathway here and what was interesting to us is in fact we did not find any defect in the PI3 kinase pathway in these patients and these were very well phenotyped, very well controlled PCOS patients. We found no significant abnormality. Now perhaps we needed to have many more patients, I don't recall at this point what it was, but it was a significant number of individuals who also underwent an FSIVGTT to assess their insulin action in vivo. So this was somewhat surprising to us. We also focused on looking at the insulin receptor. There had been some reports Andrea Dunaif and others had looked at fibroblasts and had said that perhaps the insulin receptor was abnormal, particularly serine phosphorylation which inhibits the action of the insulin receptor was overexpressed in fibroblast PCOS patients. We did a series of studies in this regard. This is again adapted from that data and we found no abnormality really in autophosphorylation of either tyrosine or serine insulin receptor. So again we were surprised but you know we have to take negative data. It's always nice when you publish positive data but you know the vast majority of the truth is actually in the negative data. We also looked at MAP kinase. As you know insulin has both a glucose and lipid metabolic effect but it also has an anabolic effect and most of the anabolic effect of insulin is mediated through that MAP kinase pathway. Lots of other pathways as well for those of you who are interested in that but nevertheless MAP kinase. Now what we found actually is that in general the MAP kinase pathway in general really no difference in p38 or junk pathway but the ERK pathway was actually abnormal. We found in fact that in this well-studied and again it's not just that we looked at a whole bunch of things but this was really definitely abnormal in this and some of these data have yet to be published. Now that's not totally unique right? This is a data from this is actually looking at I'm sorry our own data but looking at really the ERK phosphorylation and it which appears to be constitutively overexpressed and adipocytes of PCOS and you can see that it tends to be overexpressed and if you for example utilize a inhibitor of MAPK you will find that these various transcription factors will now increase. So overexpression of ERK tends to suppress GLUT4 activity and if GLUT4 expression in adipocytes and again sorry about that we're going to move to that. This has been reported elsewhere as well this is data from Andrea's laboratory looking at skeletal muscle and again you can see here that phosphoERK was overexpressed in this case phospho38 ERK so this was again corroborated by other other fields so we felt fairly confident that our finding of an overexpressed ERK in adipose tissue function was certainly a finding that merited further study. Now we then turned to the epigenetics and I explain the link in a second but there is significantly importance in the epigenetic regulation adipose tissue function and PCOS you know epigenetic regulation may influence PCOS related insulin resistance in adipose tissue function. We initially started like many people looking at methylation in adipose tissue and really found no significant differences in whole DNA methylation in the studies that we did in fat. We then turned our attention to micro RNAs and I'll explain why in a second and again many of you know that these are short non-coding RNAs involved in post-transcriptional regulation and they are known to influence many cellular functions including glucose and lipid metabolism. Now one of the things that actually interested us and was really the relationship between ERK overexpression which I had mentioned a second ago and micro RNA for example this is a particular ERK signaling oops and I'm just going backward on my thing here and we found a relationship again between micro RNAs and ERK overexpression. We looked at a number of differentially expressed micro RNAs in adipose tissue and really focused on micro RNA 93. People often ask me why micro RNA 93 and not the whole gamut and the reality is that we to understand in depth what the mechanisms may be behind the impact of micro RNA overexpression in adipose tissue dysfunction we had to focus on one of these micro RNAs. If you start studying the whole gamut it becomes a fishing expedition at least initially. We had again targeted micro RNA 93 not just because it was overexpressed in PCOS tissue as I showed you but also because it was predicted in silico to actually impact a glute 4 so it seemed to be a very good candidate for study. And again it was also helpful that micro RNA overexpression increased ERK phosphorylation in human differentiated adipocytes which again tied our previous findings with insulin signaling to this micro RNA 93 or epigenetic mechanism. So those three things right the in silico prediction that that micro RNA 93 would impact on glute 4 expression which was one of the principal findings that we find in adipose tissue the fact that it actually was overexpressed in PCOS adipose tissue and then thirdly the fact that overexpression actually also impacted on ERK which was one of the other findings we found in signaling really allowed us to target micro RNA 93 further. Now what we found in our studies was that micro RNA 93 overexpression was observed in all our PCOS patients whether they were obese or non obese and again these are moderate size studies with very good controls which is unusual usually you find sort of a smattering some of them have it some of them do not in fact in our population again not hundreds right but but certainly in the more than tens and twenties right they were observed in all of these individually and what we also found is that it was negatively associated with with whole body insulin sensitivity. Now what's interesting about micro RNA 93 is that it was elevated in both obese and non obese PCOS patients but it was also elevated in controls non PCOS controls who had insulin resistance. So it's very lucky that overexpression of micro RNA 93 isn't just specific to PCOS although it certainly impacts all of PCOS but it may also play a role in other insulin resistant mechanisms. When you look for example here in this quadrant here you look at the relationship of micro RNA expression to GLUT4 you can see that the higher micro RNA the lower the GLUT4 expression and then if you look at in vivo which is important to really relate what you're observing in in vitro to vivo you will see that there is a linear relationship between micro RNA 93 and HOMA right the higher HOMA the more insulin resistant the individuals so the more expression of micro RNA 93 the more insulin resistant these individuals just with HOMA. We studied a large number of these also with FSIVGTT. Now we did a few experiments we overexpressed micro RNA 93 in human differentiated adipocytes and found as we predicted that the overexpression of micro RNA 93 and I can't hear see here this overexpression over here was very clearly denoted suppressed the the GLUT4 expression as well as protein content so we did verify this mechanism in human differentiated adipocytes then we went and inhibited micro RNA 93 we used a micro RNA 93 inhibitor and you can see that that certainly suppressed micro RNA 93 on the left and and then led to over to an increase in GLUT4 gene expression on the right so again verifying that we did the alternative right if we actually suppressed GLUT4 micro RNA 93 we would find an increase in GLUT4. Now these are data again that are on a published but we went on to produce an overexpressed micro RNA 93 mouse model and we've again characterized it extensively but what I think is important to note is that actually as you look through the mouse model you will find that actually and I can't read from here but GLUT4 expression was suppressed as you would predict with an overexpressed mouse model insulin levels were high as you would expect in an insulin resistant animal and I can't tell what that quadrant is over here so you'll have to actually I have it here I'm totally it's testosterone and CYP 17 and both of these are increased as you would expect with a hyper androgenic model so the interesting thing about this this murine model is that we with an epigenetic hit we were able to actually reproduce both the metabolic defect that you find in PCOS and the hyper androgenic defect that you find in PCOS again going back to this incident this issue of who comes first insulin or androgens and the mice did not get any hairier so we were not able to verify the hirsutism effect of this thing and and again obviously none of you laugh so you don't get it okay you know yeah this is gone okay anyway looking here we began to try to we are currently exploring why is micro RNA 93 overexpressed in PCOS subjects and I should tell you that micro RNA 93 is expressed both in primary adipocyte cultures of PCOS as well as pre adipocyte cultured adipocytes right so they are conserved that overexpression is conserved not just in the primary tissue which would talk about the environment but also in in tissues that were where adipocytes are derived from pre adipocytes tromal fraction of adipocyte of adipose tissue we began to try to understand why is micro in a 93 overexpressed in PCOS patients we look for example at the host gene micro in a 93 this host gene is MCM 7 and in fact we were surprised the MCM 7 is actually under expressed in the tissue not overexpressed so it's very clear that the overexpression is a post transcriptional effect not related to the gene expression itself back the gene seems to be suppressed perhaps in response to the excess micro in a 93 so again this is something that surprised us right we would have expected that if it's a primary overexpression at the DNA level that MCM 7 would be overexpressed as would MCM and micro in a 93 but it was in fact opposite and then we began these are studies that actually preceded this but one of the other observations that we've had in adipose tissue and I don't have time because our chairs are going to yank me off the podium is is adipocytokine dysfunction in adipose tissue for example this is just an example of that we find that adipose tissue of PCOS patients overexpressed for example TNF alpha right and and if you for example look at the impact of TNF alpha on adipose tissue production of adipocytokine adiponectin a insulin sensitizing adipokine what you find is that in controls TNF alpha if anything may have increased adiponectin production in controls perhaps in response to the inflammatory effect of TNF alpha but in PCOS in fact TNF alpha actually went on to suppress adiponectin even further indicating that actually the adipose tissue in in PCOS patients tends to be pro-inflammatory in its response to actually other adipocytokines and there's a whole line of research here but as I said we're not going to get into that but the reason I bring this forward is that in fact when you look at the regulation of say TNF alpha and the relationship with micro RNAs you will find that in fact I'm sorry here let me get my controls that the TNF alpha tends to lead to overexpression of various micro RNAs including micro RNA 93 and in fact if you if you look at GLUT4 expression and you administer TNF alpha on human differentiated adipocytes it will suppress it you then use an inhibitor of TNF alpha and that will then restore GLUT4 so perhaps the access has to do with both the adipocytokine interaction with the epigenetic component of adipose tissue which leads to this chronic overexpression constitutive overexpression of micro RNA 93 and probably many other micro RNAs but this is the one that we've targeted for our research. So in conclusion insulin mediated glucose transport and lipolysis and I have not shown you the data on lipolysis but there was a poster that we had today as well as an oral of the androgen excess and PCOS society talking about NEFA signaling but again the the defects are not just in glucose transport but also in lipolysis and PCOS and so insulin mediated glucose transport lipolysis and adipose tissue and PCOS is dysfunctional and associated with really increased cytokine expression and dysregulated adipocytokine action fostering a pro-inflammatory milieu overexpression of micro RNA and other micro RNAs are also observed and that tends to lead to constitutive over activation of ERK possibly related to the cytokine for example TNF alpha and other micro RNA over activity and decreased GLUT4 content which may at least be in part due to the action of both the TNF alpha and other inflammatory cytokines derived from adipose tissue macrophages and the overexpression of micro RNA 93. So there seems to be a number of mechanisms impacting on the adipose tissue dysfunction the pro-inflammatory function of adipose tissue and PCOS related to again adipocytokines epigenetic changes and they actually are not entirely separate they actually tend to collaborate and they tend to overlap as well. So I think that is the the end actually micro RNA 93 is overexpressed in adipose tissue PCOS the overexpression inhibition of micro RNA 93 leads to suppression or increased GLUT4 as I showed you in the data. A micro RNA 93 overexpressed mouse model really sort of tends to replicate the PCOS like phenotype very uniformly with a single hit. The mechanisms underlying the constitutive overexpression of micro RNA 93 in adipose tissue of PCOS and perhaps individual in insulin resistant individuals remains unclear but these data do suggest that epigenetics is an important component of the adipose tissue dysfunction in these individuals and perhaps will relate to their metabolic dysfunction in vivo. So this is again a part of a continuing research in our laboratory it is exciting it's interesting but again I think that adipose tissue dysfunction plays an important role in PCOS but it also likely plays an important role in other parts of the body and other insulin resistant syndromes and it's something we probably should continue to examine closely. Thank you very much. Questions open for questions while people come to the mic I'm gonna take the prerogative of starting off. Frank I should have had some of your slides up there but they only let me have some. And you finished ahead of time Ricardo I did not think I would be uttering those words today. You put the fear of God in me. This is like I'm you know. That's for more questions. Does your does your does the adipocyte function carry through if you look at lean women with PCO as well as obese so it's it's regardless of their sort of phenotype in terms of lean or obese or metabolically healthy or unhealthy. So most of the studies that we actually have published in this are actually in lean women with PCOS and as you know people talk about obesity and PCOS and that's a whole other subject that we have studied but in in general about in the clinic about 60% of PCOS patients even in our clinics in Alabama which is not a thin state are obese and 40% are actually lean so even in the referral clinic. When you study PCOS women that are identified in more and larger population studies that we do a lot actually about 60% are actually lean. So while it's certainly a little harder to convince lean women to give you fat it's not that hard and these most of these findings are actually lean women because we wanted to control for fat. Actually Nanette was one step ahead and similar thinking so I was just going to ask you compared to your controls what was the BMI of controls versus your PCOS we heard their lean so I was just curious. Yeah there we try to match controls and PCOS women very closely so we have this kind of rubric and Fernando back there who's helping me with some of these studies will remember it but we match once we recruit our PCOS we tend to either recruit specifically for our studies or recruit from a pool of controls that we have and we try to match them plus or minus three kilos in weight plus or minus three years in age and of the same race or ethnicity so we try to be very careful sometimes we actually have you know they still tend to be a little bit younger and a little bit more heavier and so on but we tend to try to match them very carefully in that regard which is why I think we are able to get clean data because we you know clearly phenotype these individuals we also use the NIH 1990 criteria which I didn't speak about because again we want to use classic PCOS if you would and we are interested in the metabolic dysfunction it's phenotype A and B of Rotterdam in case anybody is keeping count so thank you Frank. There Aaron Sipas NIDDK NIH I want to know if you've identified any treatments or conditions that lead to a reversal of the states is that maybe miR-93 is lower or you've at least been able to re-increase GLUT4 expression in these cells. Well we submitted a few grants to the NIH and so you know let's let's see what you guys say but I'm not part of that oh I know I originally review enough don't worry no but my point is so we are looking at both nutritional aspects of intervention and its impact on adipose tissue dysfunction as well as pharmacologic and I'd rather just stop there and say that but we are looking at some of that to see how much reversibility we're able to obtain both in vivo right because we we sort of know what happens in vivo with a lot of these drugs what we don't know is what happens at the adipose tissue level and that's what we're beginning to do that but that's a great question and that's exactly our next steps. Thank you. Oleg Varlamov, OMPRC. So on the mechanism of PCOS dysfunction in adipocytes. So you state that GLUT4 expression is reduced and it's a mechanism for PCOS dysfunction right or at least for insulin mediated glucose uptake suppression yeah sure the question is I've read a few of your articles so what about recycling because most of GLUT4 is inactive so it's in the endosomes have you ever measured actual recycling or glucose uptake in fact in isolated adipocytes? So we measure glucose uptake using both treated glucose back when we could use treated glucose and then of course with chemiluminescent methods today so we do have functional evaluations of adipocytes and glucose transport in adipocytes we also have functional tests in regards to lipid right glycogen accumulation etc. We I will tell you that we tried many years ago to look at at GLUT4 translocation right because you have intracellular translocation right we could not get that experiment going and so if any of you guys have a great system for measuring GLUT4 translocation into adipocytes let me know and I'd love to hear that because we were we fail miserably and so I'm not going to present you that that data we tried. We actually know how to isolate unilocular adipocytes we can help so in fact that be great we can culture them in vitro or ex vivo. The reason I'm doing these talks is because I want brighter people than me coming with solutions to some of this stuff so I look forward to hearing from you. Thank you. Nick Webster, San Diego. So I had a thought about the origin of sort of the overexpression of micro RNA so that cluster is in an intron that's sort of in the terminus of the MCM7 gene but there's a splice variant that actually lacks a bunch of the upstream exons but splices down to the terminal one so that's still going to make that cluster micro RNAs but a whole bunch of the message is then missing which may be why you know the RNA levels might be down but you're still seeing a lot of the micro RNAs produced so it might be worth looking at some of the splice variants. I love that and I'd like to ask you to send me an email with that if you don't mind seriously speaking because I've not captured that but I do have a little bit of an issue and and we need to look at that but but as you said to be fair most of you know micro RNAs really come in family clusters right and and so the family cluster for 93 is 25 106 B and 93 the problem is that they're not all we've looked at 25 and 106 B and they're not necessarily all of them overexpressed right and so it brings us to mind as to okay this may not be a primary gene transcription issue, but a post transcription Modification, but I'd still love to hear that idea, and if you don't mind sending me an email. I'd love to hear that please I'm Sisha from Virginia, and thanks for a great talk. I was wondering what was the Source of the adipose tissue was it subcutaneous was it vertical and if they don't expect the opposite or different It's actually we wanted to we've always used the same source, and we tend to use a bit of unique unique Acts of tissue which by the way if you go to YouTube and just type in dr. Aziz fat biopsy You'll see the whole process. I've tried to put it out there Literally so people can actually use the same method. We don't knew use needle aspirations I find that that tends to rip up the cells and a lot You know we obviously have fairly large cells and PCOS patients right as you would imagine When you do needle aspiration you often end up shredding all the big cells and end up with sort of medium-sized cells and that changes the Context so we use an open biopsy through a son a son a meter and a half incision in the lower abdomen Sort of upper pubic areas we do a little bit of shaving if you would do a small incision remove fat from that area It's sort of abdominal if you would lower abdominal is still cosmetically Very acceptable to patients, and then we take out a chunk. You know it's a chunk about the size of my thumb And then we're able to really look at the architecture and preserve the cells and so on so forth as opposed to shred them so That's what we use, and I think that's that has helped us With us now. I will say that we used to use Just if you want to get really technical a lot of came with epinephrine We stopped using epinephrine and lidocaine just because we think and obviously Epinephrine does have and there are studies that I showed you earlier that it may impact on lipolysis We don't know that's the case, but we're trying to do it cleaner I'm only saying this because I'd like people to replicate some of these studies or not Using a similar technique, but thank you for that question. Thank you. Just if I can ask a second question Do you expect that the changes would be different or your findings would be different? It was if we were studying visceral adipose tissue So the answer is possibly We studied a few visceral patients initially because I'm you know I'm a reformed surgeon and so we were able to get you know visceral fat But the reality is that these type of studies have all been in subcutaneous tissue We have not really looked at visceral fat And I certainly think it's an important thing to do now We have to be a little bit careful because many of the findings that we find in polycystic ovary syndrome as relates for example insulin Singling and so on don't seem to be uniformly throughout all tissues, okay, and for example You know obviously you know there is insulin resistance and peripheral tissues But the evidence that there is insulin resistance at the ovarian theca tissue right which is that produces androgens It's actually much less and in fact it may be the opposite right I mean you I showed you I don't know if somebody was at the debate yesterday older data from John Nessler, right? Where you actually? In you know provided insulin to isolated theca cell from PCOS, and they hyper responded right to androgen So so you have to be a bit tissue specific And this is one of my concerns about looking at lots of tissues We need to do that, but I wouldn't necessarily be surprised if if it's not the same. Okay. Thank you very much Okay They're hiding me behind here. They don't want to see my face our next speaker is Dr.. Steve Stephen Frank's Dr.. Frank's was trained in internal medicine and endocrinology He is professor of reproductive endocrinology at Imperial College, London He is a former chairman of the Society for endocrinology in the UK He has both clinical and laboratory based programs of research in the field of normal and disordered function of the hypothalamic Pituitary ovarian axis, and he has a major interest in both reproductive and metabolic aspects of polycystic ovarian syndrome So I would like to welcome dr. Frank's to begin his presentation Thank you very much and good afternoon ladies and gentlemen, so Let me move on to the disclosure and then the outline of my talk Which is to first of all say something about obesity and polycystic ovary syndrome Then I'll go on to some studies of energy balance in women with PCOS then related to that talk about androgens and brown adipose tissue and finish off by Looking at ways in which we might be able to modify energy balance in polycystic ovary syndrome to treat obesity So first of all some thoughts about obesity and polycystic ovary syndrome And as Ricardo pointed out You get different prevalences of obesity whether you're looking at Population studies or clinic studies, and this is a an ongoing population study that I've been involved with for some time Which is looking at the North Finland birth coat or cohort, so this is a cohort of of Something over 5,000 women who've been studied since birth at the age of in 1966 and followed through at the ages of 14 1931 and 46 with some bits in between as well But we have full data from those time points and what you see here is that In this particular series, and these are self-reported symptoms of PCOS But you'll note here that they're actually in this particular case. They fulfill the Rotterdam the NIH criterion that they both have they have both oligomenorrhea and hyper androgenism and what you see is that about a quarter of the patients were Obese compared with just 8% of the control reference population of course there are changes with age which we all suffer from And you see this difference is actually amplified so at the age of 46 You see the majority or almost the majority of women with PCOS have have obesity so that suggests that they're more prone to be obese as they get older and And indeed what we see is that even at the age of 14 There's a predisposition to weight gain because the weight gain between the ages of 14 and 31 is Greater in the PCOS women compared with the reference population So why is that? Well, we don't really know but here's another piece of data which looks at the so-called adiposity adiposity rebound Which is this inflection in the BMI that you see? during childhood and essentially the earlier you see Adiposity rebound the more likely you are to be obese in later life and indeed what we see here is that women Who develop polycystic ovary syndrome as a child they will have an earlier adiposity rebound So how does that that might well contribute to a greater risk of obesity What else might do well, of course genetic predisposition and and recent studies using Mendelian randomization Suggests that genes which are a cause of obesity are also causally related to PCOS. There's a tight link there the developmental programming by androgen and we've already heard something about that from from Ricardo in the Epigenetics may well be involved and it may well be androgen stimulated and we'll hear more from for Santha in more detail about the effects of androgen program and Programming so there are effects on adipose tissue and the good question I think is is early adiposity rebound an androgen effect. Is this a programmed? Epigenetic effect we don't know and of course dietary intake will have a part to play as well So, let me move on now to energy balance in polycystic ovary syndrome Now when we look at daily energy expenditure the majority of Energy expenditure, of course is resting energy expenditure or basal metabolic rate, which you see in the gray exercise 10 to 15 percent and rest and interestingly postprandial or diet-induced thermogenesis accounts for a significant proportion of daily energy expenditure So on the premise that this is modifiable we looked at women with polycystic ovary syndrome and I'm pleased to announce the the 30 year anniversary of these these studies which my colleague Stephen Robinson led and what we show here is that these are women who are We've got 16 women with PCOS and 16 individually weight matched pairs BMI match pairs So they're very carefully matched and we're looking here at the increments of energy expenditure Following a mixed meal. So postprandial postprandial energy expenditure Which is in the form of heat and is measured by indirect calorimetry and what you see here is in the closed symbols is in the control subjects and in the PCOS subjects the in in the open symbols that the women with PCOS have significantly reduced postprandial energy expenditure So, of course reduced energy expenditure means an increased tendency to gain weight and There's of course one of the features of women with PCOS Is that they tend to be more insulin resistant and these are old data from the same study from Steve Robinson And of course, there are many other studies and Andrew Denef was one of the first to show Using clamp studies that there were differences between PCOS and normal How does that relate to what we see in? Terms of thermogenesis. Well, there's a inverse correlation between Thermogenesis and insulin sensitivity or rather insulin resistance. So the more insulin resistance you are The more likely you are to have reduced Thermogenesis Suggesting that insulin of course is playing a role here So is Reduce postprandial thermogenesis in women with PCOS an effect of androgen programming This was a question that we asked ourselves some time ago and then more recently in a collaboration with Colin Duncan in Edinburgh We've had a chance to look at this question So this is a Scottish blackface sheep are looking a bit gray faced there, but these are and this is the model that we've we've we've used and Let me just say a bit about that. So these are animals that are looked at at some 20 to 24 Months, so they're beginning to cycle. It's in their second breeding season They're given Testosterone proprionate 100 milligrams twice weekly and that's between days 62 and 147 of Pregnancy, so mid to late gestation in the sheep And the Santa will tell you About some data with which I think you've used an earlier window of of exposure to androgens And lo and behold we see a very familiar pattern so here is Here are the data in response. So so these animals are given a Subcutaneous thermometer which is measuring heat and they're given a standardized meal and what you see is that in the control sheep There's a nice postprandial response and it's reduced in the pre and prenatally androgenized animals So looking very very similar to what we see in women with polycystic ovary syndrome What's the reason for reduced postprandial energy expenditure Well, there are lots of factors involved in energy expenditure after eating and Let me just Give you a bit more detail of that. So if we these are the factors which we think are involved They are aberrant insulin signaling altered sympathetic activity Adipocyte dysfunction and excess androgen and they're all interrelated and I don't think they're independent of each other So, let's just look at insulin signaling and these are data from Colin Duncan's group that we've done together. So here he's taken after sacrifice of the animals he's looked at Hypothalamic implants and he's looked at the effect of insulin On these implants insulin signaling and he's looked at the map kinase pathway And the pi3 kinase pathway and you'll see that in both cases there is reduction here of Pi3 kinase and Erk in response to insulin so insulin resistance in early insulin signaling So, let's move on to so a defect in insulin signaling in these animals may contribute But what about the effect of on brown adipose tissue? Well, what we see here is The effect using these prenatally androgenized animals We're looking at fat depots from the neck from the back, but that's the inter scapula region visceral and gonadal fat and as you see in each case the expression of UCP1 uncoupling protein 1 which is involved in thermogenesis is reduced And That moves us on to some studies that have you learner in our group has has been doing in immortalized white and brown Adipocytes from from mice. So these are immortalized cell lines What I'm showing you here is the data in brown adipocytes and you'll see they're they're differentiated so that the differentiated Adipocyte Has this appearance and what we're looking at here is androgen receptor Which is abundant in these adipocytes and here looking at the effect of increasing amounts of dihydrotestosterone in vitro And you see with increasing doses. There's Abnormal morphology of these of these cells and I think we're probably killing them at that point And here is a View using oil red. Oh see not lots of lipid here, but in the presence of 10 micromolar dihydrotestosterone There's really quite a marked morphological change and these markers of differentiation are suppressed by increasing doses of androgen, so androgen having an effect on differentiation of brown adipocytes and And perhaps more importantly we see that there is an effect on mitochondrial respiration And we've used the seahorse method to to look at that And you see that this is a dose related effect in terms of mitochondrial respiration both basal and maximal respiration So again related to to thermogenesis Well What about brown fat in humans, it's most abundant in in newborn and early infancy There's less brown adipose tissue in adults But it may still be physiologically important and there are lots of work going on at the moment to look at this and A recent study showed that there was a higher proportion of brown adipose tissue In associate is associated sorry a higher proportion brown adipose tissue is associated with increased energy expenditure So this little fellow fairly newborn will have quite a lot of brown fat particularly in the in the inter scapular region Here he is at 15 months he's got a bit less and Here he is at 8 years old and he hasn't got much at all at least not until he gets into the fridge because What we see now and these are really classic studies showing that if you label fluoroxy deoxyglucose And you do PET imaging you can actually light up Brown adipose tissue in the scapular region here In both women and men so cold exposure will increase that it's there So It can be detected by imaging In adult humans, it's more prominent after cold exposure And what I was going to say earlier is the fraction of brown adipose tissue as a proportion of total body fat is lower In PCOS compared with controls. That's one study, but an interesting one Okay, so Can we do anything about it? Can we modify? Energy balance and of course what I'm talking about here is how can we improve postprandial energy expenditure so Increasing energy expenditure rather than reducing energy intake which of course is still important, but this is another approach So again we come back to the work that we've been doing with Colin Duncan and his team and This is the effect of giving Insulin intranasally, which we know will reach the hypothalamus And what you see here is that? in the feeding controls There is of these androgenized animals You get a thermogenic response. If you give insulin alone, of course, you'll you'll Reduce that you'll get that profile, but if you give Insulin to these androgenized animals After feeding you improve postprandial energy expenditure quite dramatically so Big question We'll find out this is something that we're actually in the process of Recruiting women for and we've heard at this meeting plenty of Papers where the idea of sniffing peptides is Coming into fashion, but here we think it might be useful in women with PCOS So in summary ladies and gentlemen There's an increased tendency to obesity in women with polycystic ovary syndrome Which probably involves both genetic as well as epigenetic factors and of course as dietary factors as well Energy balance is reduced in women with polycystic ovary syndrome And Androgens may be involved because they impair Differentiation function of brown adipose tissue and brown adipose tissue is involved in postprandial thermogenesis In the prenatally androgenized sheep what Colin cause a calls a clinically realistic model of Polycystic ovary syndrome of a Santa will agree with that And Insulin intranasal insulin will improve increased postprandial energy expenditure So with that I'd like to thank my my colleagues who were very much involved in this Abby Lerner and Kate Hardy from Imperial College mark Christian from Nottingham Kazia Simoninovich McRae and Colin Duncan from Edinburgh and thank you for your attention We'll now open up this presentation for any questions I was wondering with the defect in the postprandial thermogenesis is that specific for a particular nutrient type because there was a paper last year showing that there's a lipid induced thermogenesis Through induction as a GPR 3 receptor. Mm-hmm So I'm wondering is this a sort of defective response to fat in the diet or carbohydrate? The truth is we don't know because we haven't looked at Altering the diet to see what which component if any particular one is involved And has an effect I mean that's at these studies you can imagine they're not easy easy to do So I don't I don't think Colleen would be very happy with me And if I go back to him and said we try a different mix of the diet, please See what happens. I Had a quick question you postulate that androgens are mediating the effect on brown adipose tissue Are we able to see Similar findings in males male sheep with insulin resistance or with other phenotypes There is a male phenotype, but but actually and we do use for what it's worth. We've looked at adipocytes Immortalized adipocytes from from male Mice and you see a similar effect, but of course that doesn't mean it in vivo It's gonna have the same effect. So I can't really answer that question. I would expect it to have similar effects. I I don't know if there was any data correlating like androgen levels with well, yes your expression I think it would be difficult to do giving excess and you and giving anti-androgens is never very popular in men So I have a question about the insulin the intranasal insulin to the sheep How how much do you give and how do you are you making the sheep? Hypoglycemic are they getting how do you monitor a sheep for hypoglycemia? I can't now that I'm afraid remember the dose, but you don't make them hypoglycemic. They don't sort of keel over and okay They're not bumping into walls from the So no, it's and I think it's the local Injection if we like to the hypothalamus, we know it gets the hypothalamus So I think that's that's the postulated mechanism, but you don't need a huge concentration if it's if it's quite local Hi, I'm Annie Newell Fugate from Texas A&M University I'm really enjoyed your talk and the diversity of what you covered The question I have is actually related to the mitochondrial respiration I noticed that you kind of had to whack him over the head with a pretty I mean, it was like one micro molar I think that showed the actual dramatic Decrease and so my question is kind of twofold one What do you think about that and two would you do you think that you would see this? same-response with testosterone instead of DHT for that readout parameter of Respiration I suspect you would But I don't have the data to support that as for the dose. Yes. Well spotted. I mean a micromolar Testosterone. Well, you saw the effect on the morphology of the the brown adipocytes. They don't like it But there was a hundred micromolar which gave a significant Hundred nanomolar. Yeah, I saw that it just wasn't down as low right? Okay. No and and we've all But we didn't have a chance to look at the ten Nanomolar, that's what I was wondering if you look at that, but that did have an effect in terms of differentiation Okay. Yeah, so we would expect it to have an effect. It's just that we did we it wasn't possible just to Extend the experiment to look at and did you look at any other mitochondrial parameters? Yeah, besides that. Yeah We did we looked at very various So we looked at mitochondrial DNA for example, which wasn't changed, right? Okay. Yeah, there's a lot of literature on it. Thanks I'm so glad you came back to that Robinson paper because the first time I cited it You didn't seem to like it as much as you do now when I've I've always liked it So so the next question is Have you and I'd love all the different parameters that you're putting into the model now to try to tweak and refine how we understand thermogenesis and particularly postprandial Thermogenesis and resting energy expenditure and things that really go back to the core of metabolism So the question is have you looked at any? trans women to see what the effects are or trans men to see what the effects are in some of the Parameters that you've now worked out so beautifully in the lab not knowingly No, I mean, I think we would love to do further clinical studies on thermogenesis in in all sorts of conditions and and And obviously female to male Transsexual would be interesting group to to look at we haven't done that But it's something we'd very much like to do Okay, I do not think there are any questions online But let's check one one last time show my last slide. It's very bad Your special thanks slide yes, no no there was also at the end there was the slide with the What's it? Yeah, yeah, so for some reason it's not in this This is probably an earlier one should I load it up. Yeah, so I'm sorry about that So they're all those people listening All right. Thank you very much. Thank you Okay, and I will now introduce the third speaker for our symposium Dr. Vasantha Padmanabhan earned her PhD degree in cytogenetics from the Indian Institute of Science and completed postdoctoral Training in cytogenetics and reproductive endocrinology at the Ohio State and Michigan State Universities She started her faculty career at the University of Michigan where she is now professor emerita in service of pediatrics obstetrics and gynecology molecular and integrative physiology environmental health sciences and nutritional sciences She has published nearly 250 peer-reviewed papers using large animal models and a human pregnancy cohort and trained numerous clinical scientists and basic researchers Continuously funded by the NIH since 1988 her translational studies have centered on mechanisms governing developmental programming of reproductive health and metabolic deficits with maternal to fetal implications for PCOS and metabolic syndrome amplification of the PCOS phenotype by postnatal obesity transmission of PCOS across generations Lipotoxicity of peripheral sites by exposure to prenatal testosterone and the impact of gestational exposure to endocrine disruptors on offspring reproductive and cardiometabolic outcomes She's going to be talking about developmental programming adipocyte depot specific transcriptional regulation by prenatal testosterone excess in a sheep model of PCOS Thank you, and so I'm sort of petrified to follow the two PCOS gurus here coming in with the sheep translational model and so I have nothing to disclose and here is the QR code that if you need to Tap into it. So as an outline I'm going to talk a little bit about the sheep model of PCOS phenotype the prenatal testosterone Treated sheep that we work with and I'm going to touch upon insulin sensitivity and specifically at the tissue level Then go move on to the adipocyte phenotype, which is the focus of today's symposium Then I'm going to discuss a little bit on the adipose contribution to insulin sensitivity and then how the adipose depot specific Transcriptional regulation is occurring and then relate the transcriptional regulation to functional outcomes So before I proceed I don't want to forget thanking all the hard work of the post pre and postdoctoral Fellows who have been before and now in terms of doing really hard work in terms of sheep studies that go on I Specifically want to thank four people put a by a Tapa Dow Thangaraj, and Ellen Govan, who are specifically responsible for some of the adipocytes we're going to talk about. Kelly Bakulski, who is our bioinformatics collaborator. I have zero knowledge with bioinformatics. Without her, we can't analyze most of the large data set. And Dr. Dan Dumesic, who helps us bring some translational significance. And Steve and others, too, have contributed in terms of my discussion from clinical translation. So I think I don't need to go into this. This is a clinical problem, incidence of 5% to 20% of reproductive age women, the features of ovulatory dysfunction, hyperandrogenism, and polycystic ovarian morphology. Rather, I should say multi-cystic or multi-follicular ovarian morphology because this is not cysts we are going to talk about in this context. From a reproductive consequence, there are various outcomes in terms of menstrual irregularity, abnormal uterine bleeding, infertility, endometrial cancer. There are several pregnancy complications, preeclampsia, gestational diabetes, prematurity, as well as small and large for gestational age babies. Ebenezer has contributed quite a bit in that regard. Metabolic consequences, again, there is dyslipidemia, there is NAFLD, there is cardiovascular consequences, glucose intolerance, type 2 diabetes. So the whole panel of metabolic consequences they are at risk for. So in terms of metabolic defects in PCOS, possible mechanisms, in addition to systemic dyslipidemia, adipose tissue, which may stem from visceral hypertrophy or hyperplasia. I don't think we have data in there from a human side, and we have reduced subcutaneous adipocytes. Dr. Dumesic has actually contributed to that. All these, plus the liver, lipotoxicity, all these contribute to insulin resistance and the consequent hyperinsulinemia. So we use an animal model of PCOS. Sheep is a prenatal testosterone sheep model. So in terms of programming, sheep is 147 gestational day. We treat them from day 30 to 90. This spans the sexually dimorphic window, as opposed to the one Dr. Franks was talking about, where they program from day 60 to 90. So those animals are not virilized, ours are. So in a maternal sense, these animals show increased testosterone. Obviously, we are giving them, and they also show hyperinsulinemic status. At the fetal end, we actually see an increase in testosterone, as well as estrogen, possibly programming both testosterone and estrogen, possibly. And then I think I want to point out the levels of testosterone achieved in the fetus are very comparable to a normal controlled male fetus. So we're not doing any pharmacological levels. These are levels that are normally found in the male fetus. So just for comparison, these are cardiocentesis studies of Bekbekos et al. from Italy, who had shown that about 40% of the female fetuses do see levels of testosterone in the male fetus. So I don't have time to go into all the details, but I think pretty much by using this paradigm, we create almost all features of women with PCOS, and oligo and ovulation, hyperandrogenism. It's functional. The levels are so low, hard to detect, in terms of sheep leveling testosterone. They're all in the picogram range. They're hypergonadotropic. And then in addition, you do see mulch. So we see all three paceo-criteria in this sheep. Then in addition, at the metabolic side, we see insulin resistance. We see reduced adipocyte. And these animals are also hypertensive. We have shown considerably that obesity amplifies the phenotype. So this shows the prenatal testosterone effects, disrupts insulin sensitivity when we do a glucose challenge. What we see is a hyperinsulinemic status. The insulin sensitivity index is reduced, suggestive of insulin resistance. And one thing I want to point out is, these animals, both the control and prenatal animals, are weight in the sense there's no adiposity involvement. We have these animals. We weigh them on a weekly basis in the maintenance diet. So adiposity should not be a confound in any of the studies we perform. So these animals also show lipotoxicity. Their liver, these are animals. You can see these are already staining. The prenatal T shows a lot of deposits of fat. So this shows the cardiac phenotype. You see the systolic on the top and the diastolic in the bottom. What you see is a hypertensive phenotype. They're mildly hypertensive. They're not very severely hypertensive. So then coming out to the adipose tissue, which is the focus of our talk today, these are key determinants of peripheral and tissue-specific insulin sensitivity. Both excess adiposity and deficiency seems to induce insulin resistance. So both genetics, nutrition, lifestyle factors, exposures to endocrine disruptor chemicals such as those contributes to defects in adipose distribution and function. So many animal models have shown that testosterone excess during gestation support the developmental origin of adipose defects. So these are from our sheep. I think shown on the left of the panel A is the visceral adipocyte. On the right is the subcutaneous adipocytes. In the bottom is the distribution of adipocytes. You can see there's an increased proportion of smaller adipocytes both at the visceral level as well as the subcutaneous level. So in terms of distribution, you're sort of leaning towards more smaller adipocytes. So when we think about ontogeny of adipocyte and pancreas, sheep are a good model in terms of human because these are precocial models and so their development gets completed in utero just like in humans. And that's true for both adipocytes. You start seeing mature adipocyte before birth and that's true of human too. So this one shows that prenatal testosterone treatment also inhibits adipogenesis. These are adipose tissue-derived mesenchymal stem cells. These are from visceral fat. When we culture them and grow, you can see that the adipogenesis is disrupted and this is reflected in the differentiation markers which is P-power gamma, P4, as well as LPL. And as you can see, that you do affect in terms of adipogenesis which probably contributes to the large proportion of small adipocytes. So when we started looking at mediators of insulin sensitivity and thinking about one item, we are thinking about the balance of the negative and positive balance. From a negative balance, we are thinking about inflammation, oxidative stress, as well as lipid accumulation which contributes to the lipotoxicity. In terms of the positive regulators, we are dealing with adiponectin, antioxidant, as well as P-power agonist. So we are always looking at a balance of all these contributors, what the sum effect could be. So this one shows prenatal T-axis effect on visceral adipocyte gene expression. On the top panel is shown all the inflammatory markers that we have looked at, and you can see there is an increase in pretty much all the inflammatory markers. In the middle panel is the oxidative stress. The one on the left is the oxidative stress, nitrotyrosine, so that's increased. Along with that, antioxidants are also going up. SOD1, SOD, and glutathione, they're all going up. Then you come down to the bottom panel, which is the thermogenic markers, which is sort of a healthy phenotype, and you see they're also going up. So in parallel, as the negative markers are going up, the positive markers are also going up. So these are gene markers that we have looked at, gene expression changes in adipose on the right and liver on the left. We looked at all these changes. This looks at insulin signaling pathway. What we see is adipose is actually insulin-sensitive because you can see pretty much, like IRS2, PI3K, AKT, pretty much all those are upregulated, not downregulated. Contrast that with liver, where everything is downregulated. So liver is insulin-resistant. The visceral adipose tissue, in our case, is insulin-sensitive. So this was also proven at the functional level, where we did cultures of visceral adipocytes and stimulated with insulin, and you can see that phospho-AKT is going up, and it's blocked both at the liver and the muscle level, suggesting those two are insulin-resistant, but it's not true at the visceral as well as subcutaneous. Both of them are remaining insulin-sensitive. So in terms of mediators of tissue-specific changes in insulin sensitivity, in the middle when you think about adipocyte, so you are thinking about both antioxidants and thermogenic markers seems to be countering some of the inflammatory and oxidative stress markers, so making them more insulin-sensitive, so you see an increase in phospho-AKT. So contrast that with the liver, you see the opposite, so the lipotoxicity and oxidative stress markers seems to override the AMPK stimulation, which I didn't have time to show, suggesting they are resistant. That seems to be the same case even at the systemic level. So when you look at programming of lipotoxicity and insulin-resistant, the contribution of visceral fat appears to be coming from the reduced differentiation aspect, where there is a reduction in adipocyte size, obviously leading to ectopic lipid up-inclination at the liver level, and then you can also see those are also affected. All these are contributing to the lipotoxicity, which is obviously contributing to the insulin resistance. So when you think about adipose tissues, we were only working originally with visceral depots and then added subcutaneous, but there are quite a few depots. I'm only mentioning a few that we worked on. They're all different functionally. Subcutaneous adipose favors uptake and storage of free fatty acid and triglycerides. Visceral adipocyte promotes a higher rate of fatty acid turnover and lipolysis. There are other smaller depots, which are surrounding the heart, as well as the kidney, ovaries, and other areas. These obviously have specialized functions, which is associated with organs that are nearby. The lineage studies show that it seems to have different lineages as well. So when we look at visceral adipocyte, I'm going to summarize a whole bunch of data. So what we see is an increase in inflammation and oxidative stress. In parallel, we see an antioxidant increase in thermogenic adipocytes as well as adiponectin. So when you look at the insulin signaling members, you see increases in all these insulin signaling pathway as well as phospho-AKT, showing that the positive balance is outweighing the negative balance, leading to them being insulin sensitive. And this seems to be the same in terms of subcutaneous adipose tissue as well. Increase in inflammation and oxidative stress, increase in adiponectin and thermogenic adipocytes. But we do see a decrease in some of the gene expression. However, the phospho-AKT was upregulated, suggesting functionally they were insulin sensitive. Again, it suggests the positive regulators are outweighing the negative regulators. When we come to the epicardiac tissue, which is surrounding the heart, what we see again, there is increase in oxidative stress. We do see increase in adiponectin and thermogenic adipocytes. But we do see a decrease in IRS-1 and some of the insulin signaling molecules. We have not looked at phospho-AKT at this stage, so we really don't know whether it's insulin resistant or not. But there are clearly cardiac defects. There's cardiomyocyte function defects with Arpita Vyas, who is in CNSU. He's looking at the cardiac phenotype. He has a grant looking at some of the cardiac phenotype in this instance. Then when we come to the perirenal adipocyte, because here the information even more limited, it looks like they're all going up. Seems like there seems to be balance. We don't have a whole lot of information on the insulin signaling pathway leading up to both. Whether it's sensitive or resistant, we are not sure. So renal contribution to hypertension appears to be still studied. So I think all this led us to our hypothesis that prenatal testosterone excess will induce adipodipo-specific disruptions in gene expression, and gene expression pathways consistent with their phenotypic outcomes. So we obviously took the global RNA-seq approach so we can start looking at gene cascades. So this shows the four depots that we looked at, and each depot is against the other T depots. So you can see the gene expression profiles are very different in all the four depots. So one against the other three, you can see differential expressions specific to a particular depot. So now this is an expression impact of prenatal testosterone on the four depots. We see the most change at the VAC level, which is the visceral adipose, then the ECAT, which is the cardiac level, and we do see changes at the subcutaneous level, which is the SAT. The renal level, we didn't see a whole lot of changes. So this shows the number of genes differentially expressed in prenatal three in each of the depots. On the bottom set, you can see VAC shows the highest change, and then followed by ECAT, SAT, and the kidney. And then, again, you can see overlap. There's very little overlap. I think the line that's going up and down connecting suggests any kind of overlap between the different depots. So this, again, shows pathway overlap. So on the right is the heat map. You can see that the four different depots are showing very different signatures, and this is, again, reflected in pathways. The majority of the pathways affected were chromatin histone RNA, immune extracellular matrix, cytoskeleton, the Golgi mitochondrial transporter. There's a whole bunch of pathways that are affected, and they were all very specific for each depot, and they all showed a very different profile to us. So that brought us to the non-coding. There was so much emphasis on the histone and the RNA and other things, so there's obviously epigenetic changes that have occurred. So we looked at a few of the non-coding RNA, the long non-coding RNAs, which are implicated in some of the epigenetics. The microRNA obviously has been shown to be implicated in pathogenesis of PCOS. We also looked at the small nuclear RNA and the small nuclear RNA. The small nuclear RNA participate in alternate spicing, and pre-mRNA molecules. The small nuclear RNA is actually involved in mRNA editing and genomic imprinting. So we started looking at both the different non-coding RNA distribution. This shows the differential expression of non-coding RNA in the prenatal T-treated animals. You can see on the top left is the cardiac phenotype. That's where we saw a lot of changes. Then, of course, the right is the renal. We didn't see a whole lot of changes there. VAT, again, we didn't see a whole lot of change here. And then the snow RNA, we did see some of the changes. So I think these are all different. The top two and the bottom left are all microRNA, and on the right is the snow RNA, and that is what we saw a lot of changes in the VAT. So then we started relating the non-coding RNA to coding expression to see if there is any association between these three expressions. So here is MIR3955, which seems to correlate with the ITGB2 gene, which encodes an integrin beta chain and forms multiple heterodimeric targets. And these regulate white adipose tissue insulin sensitivity and brown fat thermogenesis. And this seems to have a negative regulation between the microRNA and the coding gene. Then in the heart, we saw several associations, and here we saw MIR3959 was associated with the HoxB6 gene, which is a pre-adipocyte precursor gene. Then MIR381, which was associated with TNRC6C, which is gene silencing. Then it's a positive regulator of nuclear transcribed mRNA catabolic process and mRNA poly-A tail shortening. Then MIR485LMOD1 is high in skeleton muscle. We haven't found an association with any of the adipocytes. These are all ongoing studies. They're not complete. Then we saw, as I mentioned, we saw quite a bit of association with the snow RNA and the coding RNA correlation. So one of the snow RNA was increased the TRDN, so it suggests there is an up-regulation in adipose tissue of insulin-resistant individuals. Another second is the snow RNA. It's a soluble protein of the endoplasmic reticulum, and it's associated with BMI in gluteal tissue and is overexpressed in obese type T2D rat model. Then the third one was associated with LMTK3, which is involved in phosphorylation of estrogen receptor, and all of us know the estrogen facilitates adipocyte deposition and function. So to summarize all these data, in an adipose depot-specific transcriptional phenotype, just looking at the control, what we saw with the epicardial fat and the renal depots, what we saw was increase in cardiac and renal functional genes, and in the visceral, there was increase in VAD differentiation genes. In the visceral, I'm sorry, it was adiposity and inflammation-related genes. In the subcutaneous, it was the VAD differentiation genes. Now, summing up all the effect of the prenatal testosterone treatment effect on adipose programming, so prenatal testosterone excess has had considerable effect in terms of the several of the non-coding RNAs. We do see changes in non-coding RNA. We haven't completed the analysis on those. There are quite a few microRNAs, again, tissue-specific, no RNA, and small RNA. So these probably contributes to the transcriptional changes. At the visceral fat level, which is at the bottom right, it shows decrease in adipocyte differentiation genes, increase in chromatin and mitochondrial pathways, increase in extracellular matrix genes, and disruption of inflammation-related genes. All of these seems to contribute to a decrease in adipocytes and may be obviously leading to contribution to lipotoxicity at the liver level and the insulin-resistant state. If you look at the subcutaneous fat on the bottom left, there is increase in adipose differentiation genes, and again, that's contributing to the decrease in adipocyte size. And then when you look at the cardiac muscle in the cardiac fat, it increase in cardiac muscle function-related genes and ER to Golgi transport genes. So this is obviously, in our model, it causes left ventricular hypertrophy and hypertension. The left ventricular hypertrophy is something Dr. Arpita Vyas has published quite a bit on, and she's really focusing on that. The perirenal fat, we see increase in vascular gene and extracellular matrix, cytoskeletal gene pathway. So we don't know the renal contribution to hypertension. That's an area that we are studying. So obviously, there seems to be non-coding RNA changes which precede the coding RNA changes, which those kind of changes that are occurring in the coding RNA is consistent with some of the phenotypic changes we are observing. So I think the take-home message that I do want to say is we all are focusing on sub-Q in the human. You obviously can't get any of this that we are doing in the shape and whether you will even be translatable from a human standpoint. But I think one other thing that we do need to remember is there are tissue-specific regulation that's going on at various levels, and then we are thinking about microRNA and any of these non-coding things. We have to expand our horizon to more than many of the small. The pi RNAs are coming into focus quite a bit, so I think there's a lot of things that we need to think about. Thank you. Okay, we have time for questions. And I see no questions online, but one of the things I couldn't help but notice, so I'm going to think a good thing for getting up, Ricardo, because you're probably going to ask this of you too, is I do not see overlap in the microRNAs that you've looked at. So MIR93 doesn't seem to be on the radar, given that it's a different animal and there may be other reasons, but why do you guys think that might be so? So there's about 3,000, it's over about 3,500 are above microRNA identified in the human microRNA set. In the rodent, there's close to about 103, I don't remember exactly, somewhere 3,400. In the sheep, there are very few, only about 150 have been identified. It doesn't mean microRNA93 doesn't exist here because we don't have the information to compare. The other thing is microRNA93 that Ricardo pointed out is form of a group of cluster which includes MIR25 and MIR106B. We do see changes in MIR25 in our studies, which we haven't related to coding, and that's why we didn't talk about it. Amazing work, as always, with Santa. I have two questions. One is about the epicardial fat genes that are, I guess, upregulated. So you talk about cardiac muscle function, and then it's talking about hypertension. So epicardial fat, did you actually analyze the fat genes around the coronaries specifically, or... That was the first question. We just took just what is surrounding the heart, and the epicardial portion is what we took. And then the second question I have regarding that is, did you see any changes in, not necessarily stereogenesis, but things that are related to the RAS system, the renin-angiotensin system, aldosterone system in that fat depot? I know you have it generalized here. We have to go back and look at that in more depth, because this is just getting analyzed right now. All this is pretty much unpublished data. We have looked at the coding publication. We have looked at the non-coding aspects of it. Okay, yeah. I think the cardiac work is something that Arpita is pursuing, so I think she probably will have data. She's doing quite an intensive set of studies looking at fetal levels and other levels, looking at some of those distributions. I'm sure in the next few months she'll have a few publications out. Super. The other question I had is regarding the mitochondrial pathways and the visceral fat. Can you comment more in detail on what mitochondrial function genes were upregulated? I can't really pinpoint those. As I said, it's a whole global pathways. We have looked at some mitochondrial copy number and other things at an RT-PCR level, so we saw some of those changes there. Okay. Thanks very much. Thank you, Vasantha, for that talk. I noticed that I'm slightly surprised to see there was a positive relationship between the androgenization and adiponectin levels of actually stimulation of adiponectin, which is not really what you see in clinical PCOS. There seems to be a reciprocal relationship between androgens and adiponectin in women with PCOS. Were you surprised by those findings? No. I think in the adipose-depose-specific manner, adiponectin was increased. But one of the things we were also looking at, when you're looking at adiponectin, you've got to look at the high molecular and the low molecular weight, right? I think in the clinical scenario, you're only looking at the overall adiponectin. So we were actually fractionating them to look at the low and the high molecular weight. I was a little surprised. When I got in, I actually thought adipose would be resistant. And then I realized this is possibly because our model, we don't let the animals get obese. And I think in, I don't know whether in Colin's model, are the prenatal T a little heavier than the control? I have no idea because I didn't see the data here, so I have no idea what that is. Because most of the animal models that have worked with prenatal T excess, whether it's monkey or rodents, all the animals are obese, in the sense they're overweight compared to control. So there is an obesity confound in all of that. So I'm not sure with the adiponectin story, Steve, I think it's something to be sorted out. We had got the antibody and we actually had to, we couldn't measure adiponectin per se, so those are all tissue level studies we have to fractionate. Did you look at adipo R1 and R2, the receptors? Yes, we did. I can't tell you what happened. I can't remember. In PCOS tissue you seem to have a lower expression. What did surprise me was the fact that they were, I don't know whether the phenotype is more what Dan Domasik is seeing in terms of the normal PCO weight range women versus obese women. And one of the things I thought, I don't know if Ricardo left, I think if he wants to comment, he said all his studies, even the controls, were insulin resistant. So he has an insulin resistant phenotype over all that. So is the phenotype reflection of insulin resistance or is it PCOS? That's what I wanted to ask him, but I think he left. Well, yeah, I mean they are all lean. They're lean, but they're all insulin resistant. Arguably all, yeah, I mean that's an age-old argument with PCO. Can you have PCO and not be insulin resistant? I would say no, but other people would argue with me. But he also said even in the control, the MIR 93 was elevated. So I was just puzzled how to reconcile that. All right, if there's no further questions, we will end today's talk. Thank you all so much. And thank the speakers. Wonderful session. Thank you all speakers.

adipocyte insulin resistance

PCOS

glucose disposal

microRNAs

dysfunction of adipose tissue

microRNA 93

epigenetics

obesity

energy balance

insulin resistance

prenatal testosterone excess

lipotoxicity