Isn't it wonderful to be all back here in person? My name is Arun Sanyal, and on behalf of myself and my co-moderator, Professor Tatiana Kisaleva from the Department of Surgery at UCSD, it is our pleasure to welcome you to this session this morning on liver cell, metabolic reprogramming of liver cells as a driving force of liver fibrosis. This is a combined session from the liver cell biology and the liver fibrosis special interest groups of the AASLD, and we are indebted to Yuri Popov and Natalia for helping set this up. I will now turn this over to Professor Kisaleva. It is my great pleasure to introduce Professor Fiona Oakley from Newcastle University, and the title of her talk is going to be Energy Dependent Epithelial Programming in Liver Fibrosis. I think we have a little disconnect with the presentation that's the next talk. We can switch sequence if it's easier. It may be less stressful on all parties at the end of the day. So in this case, I would introduce my co-moderator. We're off to a great start. This is a good omen, actually. You know, when things happen like this, you know the rest of the meeting is going to be great. My co-moderator, Professor Arun Sanyal from Virginia Commonwealth University, will present the talk, Metabolic, Transglutamic, and Molecular Signature of PA3-Mediated Acceleration of Steadic Hepatitis. Thank you for being so flexible. Thank you, and thank you to everyone for bearing with us. So here are my disclosures. So the PNPLA3 gene mutation was the first bona fide NASH-related gene mutation and was identified by the landmark work of Stefano Romeo and Professor Helen Hobbs. So the variant we are talking about is a I148M variant, and this is a loss of function of PNPLA3, which is a triglyceride hydrolase. So imagine you have a fat globule in the hepatocyte, and for it to be mobilized, you need to hydrolyze and get it out, and this is one of the key proteins. So it's associated with some loss of function. And this mutation is associated with steatohepatitis, particularly in obese individuals, and increases the risk of cirrhosis and hepatocellular carcinoma in several types of chronic liver disease. Chronic overexpression of PNPLA3, this mutation, in mice has been shown to lead to hepatic steatosis. Overexpression followed by silencing in a C57 mouse has actually shown to reduce inflammation and steatosis as well. And very recently, human hepatocytes have been engrafted into these chimeric immunosuppressed mice and shown to develop steatosis and induce an inflammatory response. So this is all great, and we have a reasonable understanding that if you have a loss of function of a protein that's on the fat globule and allows that fat globule to be mobilized, why you would tend to accumulate more fat. And it's also been shown that really associated with proteasomal dysfunction, you have accumulation of this protein when the expression of this gene is increased. In human studies, Dr. Yuki Jarvinian's group has shown furthermore that the PNPLA3 individuals who have this mutation, in their circulating triglycerides in the VLDL, there's an increase in saturated fatty acids. But within the liver over here, there is an impairment where they actually show an enrichment of more unsaturated fatty acids. And this leads to accumulation of, they hypothesize, increased accumulation of unsaturated fatty acids within the lipid droplets in the liver. Now all of these are giving us insights into how steatosis develops in those who have this mutant, but really hasn't addressed the core question of why is it that these people get steatohepatitis? What's that transition look like? And how does this contribute to progression to cirrhosis? How does this contribute to hepatocellular carcinoma? So what I would like to share with you today is some work that we published last year in hepatology, and is still ongoing. And this really comes back to our mouse model that we developed several years ago, which is a diet-induced animal model of NAFLD, which we thought was very cutely called Diamond. But anyway, we compared, as you can see over here, if you look at human hepatocytes in NASH versus this mouse, they do develop these balloon cells with even the malary bodies. And there is a negative staining for CK18 in these balloon cells. So it does recapitulate some of the histology of human disease. And these mice can get fairly significant fibrosis. You can see pretty good bridging fibrosis over here. And so in this model, when you start a high-fat diet with ad-lib sugar water, by four to eight weeks you get a pretty well-defined fatty liver. We just haven't looked earlier. I suspect it could happen even earlier. Progressive stooze NASH by 16 weeks, and then progressive fibrosis to about stage three by between 24 and 32 weeks. And then in the later time points, 36 weeks onwards, we have seen development of hepatocellular carcinoma, predominantly of the molecular subtype S1, S2, as described by Eugene Hoshida for the humans. And so there are many aspects of human disease that this model recapitulates. So we thought, well, that's very interesting that maybe the reason in the C57 mice experiments we haven't got a good phenotype is for whatever reason the C57 pure mouse cannot express that phenotype of NASH, et cetera. And in this setting, because we know things follow a certain sequence, if you hypothesize that if you can overexpress the mutant PNPLA3 in the mouse liver, you would accelerate the disease process. And if you can show that, then you could potentially study the disease process and understand, you know, what's going on. So if you, for example, have two mice, one a diamond mouse, the other a diamond mouse expressing PNPLA3 in the liver, at four to eight weeks, the diamond mouse would have on high-fat diet fatty liver. But with the PNPLA3, you might expect steatohepatitis and maybe some fibrosis. So we said, OK, let's test that. And so, of course, the diamond mouse is an isogenic inbred C57 S129 mouse. And we had three conditions. And because we are poor, we used AAV. We didn't, you know, generate all the knockout models. If study section was kinder to us, then maybe we would. But be that as it may, that's a different story. So here we used AAV-directed targeting with MT vector, with the wild-type PNPLA3 and mutant. And then for each, we have two conditions, chow diet, Western diet, followed them out to eight weeks. OK. So first we want to, because this was a 15-minute presentation, I didn't show you the actual data showing that, yes, it's only in the liver. It was not there in the other organs. But if you, first thing we wanted to really look after confirming that we did have liver-specific expression is that, do we have a phenotype? So this is the diamond mouse on a high-fat sugar water diet. You can see some fat, really not much inflammation, nothing much else going on. Whereas, as you can see over here, you know, there are foci of inflammation. You can see there's some perinuclear condensation of cytoplasm right around here. And so it did develop some steatohepatitis. When you look at the fibrosis, this is the diamond mouse, there's really no fibrosis. You're beginning to see some fibrosis over here in the PNPLA3 mouse. But then we said, OK, we need to understand if this is just a random thing or is there some specificity to this. So we did another follow-up experiment where we took these mice who had the PNPLA3 in it and midway, at four-week time point, we silenced the PNPLA3. And we were able to recapitulate, sort of reverse the fibrosis that we saw. So this gave us confidence that we were actually seeing some specific phenotypic changes that allowed us then to say, OK, now let's see what's going on under the hood, so to speak. So we started by looking at metabolic reprogramming because PNPLA3 is a metabolic enzyme. And so we thought we should start with metabolism. And sure enough, as you can see over here, under chow diet conditions, really not much difference. Interestingly, over here, when you look for the triglycerides, this is DAGs and these are TAGs, with the wild type, there is actually a decrease in triglycerides under chow diet. But under high-fat diet, you can clearly see that with the mutant, you have an increase in triglycerides and diacylglycerol. And when we looked at the saturation of the different triglyceride and diacylglycerol species shown in these red arrows over here, you are seeing progressive enrichment. You are seeing a significant enrichment of unsaturated species of fatty acids associated with these triglycerides. So then looking at the actual free fatty acids, what we find here, these are just the unsaturated fatty acids. Many of them were decreased, although they didn't reach statistical significance, except for 22, 6, and 3 over here. So it is basically remodeling the triglyceride by moving some of these PUFAs onto the triglyceride. The other big metabolic change that we noticed was an increase in total ceramides, and there were specific species of ceramides that appeared to be increased based on metabolomic analyses. To understand this, we looked at all of the sort of important genes, gene expression involved in ceramide synthesis, and many of them were actually increased. And then we silenced the PNPLA3, and we were able to see that actually several of these actually went back down, along with a concomitant decrease in ceramides. So this, again, seems to be, at least in this model, a relatively specific effect. So now having some insights into the metabolic milieu that changes once you play with this PNPLA3 in this model, then we wanted to see if we could connect the dots further to metabo-inflammation and fibrosis, given that there is growing sort of evidence that NASH is really a disease of accelerated aging, and there are many, many similarities to the metabo-inflammation and fibrosis you see in the heart, in the blood vessels, and in many other organs who have metabolic syndrome. So we went to a transcriptomic data that we had, and what you see are some of these KEGG pathways. They do a pathway analysis. These are the number of targets per pathway that were significantly altered. You have their p-value and the false discovery listed over here. And you can see that many of these pathways had multiple targets that were involved. Now they're not all unique targets. Some of these show up in multiple pathways, so I don't want to give you the idea that there are hundreds and hundreds of things that are changed. So because we had previously shown, about 10, 15 years ago now, that the unfolded protein response is disordered in NASH, and we thought we should look at this, and of course we see over here in this last black bar, I draw your attention, ATF4, the GRP78, and the CHOP. These are all increased. And then when we look at alarm pathway activation, there is activation of phospho-junk over here. So junk, of course, is an important inflammatory pathway in the liver. Something that really got our attention is this activation of the JAK-STAT pathway. We looked at many, all the usual suspects, right? So I'm just showing you some of the ones that really popped out. So we saw over here the GP130, we have the JAK1 and the JAK2 pathways that are sort of increased. And the particular signature for phospho-STAT3 really jumped out at us and really caught our attention. We never got a good antibody to work for STAT1, at least in our labs, so I can't show you that data. But when we look at it at a transcriptomic level, and we did network analysis, there was a pretty strong signature for upregulation of STAT1 at a transcriptomic level with all the caveats that go with just looking at transcriptomic data. But it did connect to a number of these targets, and the ones that you see with the little red ball are the ones that were at least significantly altered meeting our FDR requirements in network analysis. So there's a pretty broad footprint emanating from STAT1, at least at a transcriptomic level. And certainly, again, at STAT3 also, we find that there's an increase in genes associated with ceramides, serum-associated amyloids, angiotensinogen, PPAR-alpha, et cetera, over here. So these pathways are fairly, you know, promiscuous, so they interact with a lot of different other pathways. And so it seems there is a significant downstream impact that was linked to this. So then we wanted to also know whether this was, again, specific. So we decided, again, to go back to our silencing experiment, and this shows that we were able to knock down the PNPLA3 after it was expressed. And then we see over here that while the animals continued on the high-fat diet, you know, the phospho-STAT3, again, over here in the PNPLA3, they are decreased. So there was, again, specificity to this. Remember, phospho-STAT3 is an important oncogenic driver as well. So then we wanted to really get further to see what inflammatory pathways. So we looked at cell stress. We've looked at metabolic. Then we looked at cell stress. Now we're looking at the inflammatory signature, and again, coming back to our KEGG pathways, whole bunch of these pathways were activated. Now what was interesting to us is when we looked at the common elements through these different pathways, many of these were actually are known to be linked to the JAK-STAT signaling pathway. So we did sort of a deeper dive, combining metabolomic and transcriptomic analyses. And what we see over here is that there's this, from the sphingolipids, from sphingosine, there is actually, you know, all of these things, there you see a red arrow, were significantly increased at a transcriptomic level. And of course, this is the, these are the metabolites that were significantly increased. So this multi-omic integrated analysis further showed a strong signature for sphingosine-related signaling. That, of course, impacts many different inflammatory pathways. And then that brings us to fibrosis, where we also found that it was increase in collagen 1 and 3. Alpha-sma was increased over here. And interestingly, we also saw increased TGF-beta expression. We did do a set of experiments over here where we took condition, because we know PNPLA3 is also expressed in stellate cells. And so we checked, and our PNPLA3 expression was primarily in the hepatocytes. But to further see whether hepatocyte PNPLA3 could modulate stellate cell, what we did was we fed the, you know, hepatocytes a lot of sugar. And HEBG2, incidentally, has a mutant PNPLA3 in it. So that is a well-known trigger for increasing PNPLA3 expression. And that, using condition media from these mice on LX cells, we were able to then show that there was an increase in collagen mRNA in these cells. So putting these together, then, we start building a model where you have the mutant over here. You have increased oxidized fatty acids, DAGs. And so we think serine and availability of palmitate feeds the ceramide synthetic pathway. And then ceramides, of course, and the unfolded protein response from this metabolic stress can lead to activation of inflammatory pathways, such as junk and then FOSTAT3. Ceramides also are known to be immune activators. Together, I think this leads to an immune inflammatory signature, which is not just innate immune system, but also has components of the adaptive system. Together, they converge on the stellate cells to produce fibrosis. Now, is this a perfect story? Probably no. There never is. So we cannot precisely control the PNPLA3 expression level with our system. We don't really yet fully know if all cells get transfected. So we are actually now beginning to do some single-cell nucleus work to see what proportion of the heparocytes actually are showing this expression. It is also possible that independent of everything I've shown you, having the PNPLA3 in the stellate cells plays an independent pro-fibrogenic role, which has nothing to do with anything that I've just shown you. And then, of course, we need much more clarity on the inflammatory cell milieu, which we hope through some of our ongoing work we'll be able to show down the road. Now we're also beginning to leverage this for some other things. So at this meeting, we are now looking at this triangle between obesity, alcohol, and PNPLA3. And we have a poster here. It's poster number 3120 on Sunday. Please come by. This just shows acceleration of the injury with alcohol on top of what I've just shown you. And you can see this inflammatory response that develops very specifically with alcohol. So we think we're making some progress in generating the next generation of models for alcoholic hepatitis. And we thought very interestingly, if we are putting the PNPLA3 specifically in the liver, we've already checked it's not in the heart, these mice we know also get some diastolic dysfunction. To test the idea whether NASH acceleration can affect myocardial diastolic dysfunction is very hard to do because anything you do systemically changes the systemic milieu. So you can't figure out whether it's because of the altered systemic milieu or it's something in the liver. So we thought this would be a cute way to actually accelerate the disease in the liver without affecting the systemic milieu, which it does not, and then to be able to see what happens. So this is also a poster done by actually a college student, John Min. This is poster 2345. And this shows that when you accelerate in this PMPLA-3 mice, the isovolumetric relaxation time of the heart, the left ventricular and diastolic pressures both go up, whereas the ejection fraction doesn't change, indicating that the liver can directly somehow, we now are working on all the details, and you'll come to this poster, you'll see some interesting stuff there, that the liver can't control the heart. So we're coming back to the idea the heart is not the boss, the liver controls everything. So I think that's a good place to stop. I'm going to summarize by saying what I've shown you in our mouse is that liver-specific introduction of PMPLA-3 followed by high-fat sugar water diet led to accelerated steatohepatitis and fibrosis, increased TAGS-DAGS with higher unsaturated fatty acid content and depletion of PUFAs, along with increase in ceramides. Activation of cell stress pathways and innate immune system with a strong signature for STAT3, and fibrogenic activation of stellate cells. And we are leveraging this model to understand liver-heart crosstalk and the role of PMPLA-3 in alcohol-associated liver disease. Thank you. And these are all the people in the lab who really were critical to get this all done. Thank you so much. We didn't have questions at the end? Yeah, we will take questions at the end of the panel discussion. So we move on to the next? Yes. So it's a little bit of a surprise for us since the order we have is not the order we have followed. We go to number three, down, further down. Next, lower down, that one. Okay, so I will introduce Fiona Oakley again. Thank you. Energy-dependent epithelial reprogramming in liver fibrosis. The mouse works. Oh, here we go. Sorry. So if you just, okay, yeah, I just, you do that, and when you click the left click. Oh, got it. Thank you. Okay, now we've got the technical issues. Sorry. It's okay. Energy is sorted out. A real hard act to follow with Professor Sanyal. So thank you to everybody, to ASLD for inviting me to speak here. So this is just a little bit about me. These are my disclosures. Can everybody hear Professor Oakley? No. Sorry. Sorry. I will speak up. So those are my disclosures. And I wanted to start by addressing how hepatocytes modulate their phenotype in response to tissue injury. So when the liver is injured, the hepatocytes kind of have two fate choices. So they can either die, and this will release lots of alarmins and damage-associated molecular patterns, which can lead to tissue inflammation and drive activation of myofibroblasts and can promote fibrogenesis. Or these cells can repair, and they can adapt to these damaged stimuli in the tissue. And there was a really beautiful paper by Loft et al., published last year in Cell Metabolism, that tracked the hepatocytes using the Enact Mouse technology, where they could label the nuclear membrane of the hepatocytes, and then challenge the mice, either with a chow diet or a Western diet, and then look at how these hepatocytes remodel their whole transcriptome and their epigenome using single-cell-seq and ATAC-seq. And they show that transcription factors really cooperate together to remodel this transcriptome and mount an adaptive response. So we're really interested to understand how cellular crosstalk in the fibrotic niche is fueled in some of the mechanisms underpinning this. So we know that the transition from this healthy liver to the fibrotic liver is an energy-dependent process. If you're going to remodel these transcriptomes, increase your protein synthesis, and change the behavior of the hepatocytes, then this is going to require energy. So to investigate this, we used the C-REL knockout mouse as a tool system. And the rationale for this is that if you injure wild-type and C-REL global knockout mice with CCL4, that these mice, the C-REL knockouts, will develop less fibrosis. And this occurs in the liver, but it also occurs in other organs, such as the kidney, lung, and heart. And if you look for C-REL expression levels in the normal human liver, but then the diseased human liver, so this is from alcoholic liver disease, we see an increase in nuclear C-REL expression in the liver. And C-REL is one of the transcriptional subunits of the NF-kappaB transcription factor. So what we wanted to do is we isolated hepatocytes from either wild-type or C-REL knockout mice, and then we challenged them with a fibrogenic stimuli, so TGF-beta. And what we can see here in blue is that the wild-type hepatocytes, in response to a TGF-beta challenge, increase a number of pro-fibrogenic factors, so they actually secrete pro-fibrogenic proteins, such as CTGF, cathepsins, and both bone morphogenic proteins. But what was interesting was that the C-REL null mice, the hepatocytes isolated from these mice, retained a more epithelial biology, so they continue to secrete albumin, they continue to make ceruloplasmin and alpha-1 antitrypsin. And then if you do a targeted inflammatory screen on the secretomes of these hepatocytes after challenge, what you can see is that the wild-type mice upregulate a number of pro-inflammatory cytokines and chemokines, but this was impaired or diminished in the C-REL knockout mice when you stimulate them. So what's happening here at the metabolic level when you challenge hepatocytes with TGF-beta? We can see with the wild-type mice, which have that inflammatory and fibrogenic response, there's an increase in glycolytic rate. However, this is suppressed in the C-REL knockout hepatocytes. And so why is this? Well, C-REL is a direct transcriptional regulator of PFKFB1 and also PFKFB3. And PFKFB3 promotes the production of fructose 2,6-bisphosphate, which becomes an allosteric activator of PFK1 and accelerates glycolysis in the cell. So in response to tissue challenges, an increase in glycolytic rate in the wild-type cells. So is PFKFB3 important in liver fibrosis? So we stained control olive oil or acute CCL4 injured wild-type mice for PFKFB3. And you can see here around the central vein that the damaged or stressed hepatocytes increase the levels of PFKFB3 expression. So we then wanted to ask, what happens if you delete PFKFB3 specifically in epithelial cells and then initiate an acute wound healing response? And to do this, we took PFKFB3 knockout mice and administered an AAV expressing CRE under the TBG promoter to make it hepatocyte-specific. And when you do this and challenge the mice, we can see here in the blue bars that the PFKFB3 floxed mice have a recruitment of neutrophils and macrophages to the liver. But this is suppressed when you delete PFKFB3 and reduce the glycolytic rate in the hepatocytes. This is also associated with a reduction in the activation of myofibroblasts in the liver. So if you delete PFKFB3 in the hepatocytes, there's reduced activation of these myofibroblasts, shown by this nice alpha-smooth muscle actin staining. So how is this driving fibrogenesis in the liver tissue? So what we know is that when you challenge the hepatocyte with a fibrogenic stimulus such as TGF, then you activate C-REL. This can then upregulate expression of PFKFB3. And this causes the hepatocyte to release these pro-fibrogenic factors such as CTGF. Interestingly, C-REL also regulates the expression of PFKFB1 and 3 in macrophages, which is required for their polarization. So you've got a sort of two-pronged hit here in that C-REL in macrophages is important for allowing these macrophages to polarize and initiate inflammatory responses and activate myofibroblasts. So to test this, we took quiescent hepatic stellate cells from mice, and we challenged them with a conditioned media of hepatocytes from either wild-type at the top here, so C-REL knockout mice, or macrophages after M1 and M2 polarization. And what you can see is that actually the conditioned media from the TGF-beta1 treated wild-type hepatocytes or the M2 polarized macrophages actually accelerates the rate of which these myofibroblasts become activated in the liver tissue. So this is all in sort of cell culture systems, and it's in sort of hepatocyte models, acute models. Does this translate to a fibrogenic response in a chronic model? So what we can see here is that if you perform the chronic CCL4 model, there are flux mice here in gray. You see a nice induction of fibrosis in the liver. This is decreased in both the hepatocyte-specific knockout of C-REL and also the myeloid-specific knockout. So if you take out C-REL in the macrophage lineage, but interestingly, when you delete C-REL in both lines by giving the AAV to lysame pre-REL knockout mice, you can delete C-REL in both cell phenotypes, and this led to a greater reduction in fibrosis, suggesting some synergy there. So does this translate to a human tissue system? And to test this, we took human tissue slices and cultured them in the presence of TGF-beta for up to 96 hours and then looked at the deposition of fibrosis and activation of the myofibroblasts. And we can see quite nicely that TGF-beta challenge of these human tissue slices and culture activates deposition of fibrogenesis and the activation of the myofibroblasts. But if you give a compound called IT603, which is an inhibitor of C-REL, then we see a reduction in the fibrosis and myofibroblast activation. So I just want to switch tack a little bit here and tell you about some other recent work from the group. And this was just asking how the fed state could maybe alter how epithelial cells behave. So under nutrient-rich conditions, we get an uptake of lipids in the liver, but really what are the mechanisms that tip the balance between lipid droplet formation and lipid degradation between a fatty liver and a healthy liver? And one of those is lipophagy. So just to give you a bit of background here, so lipophagy is the specific degradation of lipid droplets, and it's kind of an autophagy process. So in the fed state, hepatocytes can form these lipid droplets here. And what we've been able to show is that these lipid droplets are coated with perilypins. So in chaperone-mediated autophagy, then perilypin 2 is important for degradation of lipid droplets in the unfed state. But then in the fed state, we see that these perilypins or perilypin 3 can coat the lipid droplets, and that in this fed state, you get activation of mTOR. mTOR can then drive the phosphorylation of plinth 3, and this brings in the autophagy machinery, leads to the formation of these autophagosomes, and then fusion with the lysosome, which allows these lipid droplets to be degraded. This can then lead to the production of fatty acids, which can be then metabolized through the mitochondria to produce ATP. But if you block mTOR signaling in hepatocytes, then actually you can block this formation of these autophagosomes and the degradation of lipids. And this was seen here, if you look at the green bars at the end, that if you block autophagy and stimulate hepatocytes with oleic acid, then you see an induction or a formation of triglycerides in those cells. And if you block the lysosome, then there's no further additive or synergistic effect, suggesting it's a common pathway. So we really know that this pathway is also going through plinth 3, and then this leads to a decrease in mitochondrial respiration here. So we can see oleic acid at the top, and then if you inhibit the lysosome or inhibit mTOR signaling using rapamycin, we see a decrease in mitochondrial respiration after oleic acid challenge. So we know this is a plinth 3 dependent mechanism. So if you give siRNA to plinth 3 to wild-type mice and make tissue slices from these and challenge them with oleic acid, we can see that we can block phosphorylation of plinth 3. We also block formation of these autophagosomes, and this leads to a decrease in fatty acids again here. The green bars at the end, so if you treat these tissue slices from the plinth 3 knockdown animals, you see an increase in triglyceride deposition, and this again is comparable with lysosomal inhibitor, suggesting a common pathway of action. And similar to blocking mTOR signaling, you see less fatty acid degradation and a reduction in the mitochondrial respiration rate compared to just oleic acid alone. So again, is this something that transfers into a human tissue slice system? So to test this, we created human tissue slices and then pre-treated with rapamycin, then stimulated the cells for 24 hours with oleic acid, and at the final two hours prior to harvesting the tissues, treated plus or minus the lysosomal inhibitor. And what we were able to see, similar to the cell culture system, is that blocking mTOR signaling in the presence of oleic acid led to an increase in triglycerides in the liver tissue, and this was comparable to using the combination with lysosomal inhibitor, again suggesting a common pathway of action. So what was the consequence in terms of lipid toxicity of these experiments? Well, if you challenge the tissue slices with oleic acid, we see a small release in IL-8, but if you block lipophagy pathways, then you see an increase in the secretion of IL-8 from these tissue slices. And we also see an increase in the production and the secretion of TYMP1, so a pro-fibrogenic factor, so suggesting that if you can't activate lipophagy, then you can develop hepatotoxic effects such as inflammation and fibrogenesis. So I guess the key takeaway messages are that hepatocytes, when they become stressed, they initiate a program of gene transcription and protein synthesis to try and adapt to this tissue injury. This requires the cells to be fueled, and one of the mechanisms by which the hepatocytes can fuel this is through increasing their glycolytic rate, and this can act as an initiating force for driving fibrogenesis and inflammation. And then for the lipophagy story, it just shows that mTORC can directly regulate PLIN3 and promote lipophagy through interaction with the autophagosome, and that stimulating lipophagy in the fed state could be a way to help protect from liver toxicity by promoting lipid droplet degradation. Thank you. Thank you. Thank you so much. Thank you. We'll move on to our third speaker, and the third speaker is going to be Dr. Steinberg, who is going to talk about AMP kinase and the GDF15 axis in NASH. Sorry. Sorry, it's a little bit of a Russian roulette over here this morning. It's because I'm Russian. It's because I'm Russian. Oh, sorry. Was I supposed to go? Wasn't there someone before me? Listen, I was supposed to be the second speaker, so I'm following whatever they're putting up over here. So I'm being surprised too. Okay, sure. So that's why that's my poorly chosen. All right. The third surprise is that I changed my title slightly. I hope that's okay. Okay. To talk about some newer data unpublished. Fantastic. Some other aspects that I think are more interesting. So thanks for the invitation. Here are my disclosures related to this presentation and that are also in the program. We're all aware of the consequences of NASH and fatty liver disease. And based on the beautiful work of Elizabeth Parks and others, we know, you know, the key drivers of these phenotype initial insults are really elevations in de novo lipogenesis, as well as, you know, insulin-resistant adipose tissue and excess lipolysis, delivering fatty acids to the liver, increasing triglyceride synthesis. And my talk will touch on both of these concepts here today and sort of try to see how models have developed that may lean more towards one or the other. And so with respect to the adipose tissue side of things, we know that the mouse has many different types of adipose tissue, flavors of adipose tissue. We have the brown fat shown here. We have the beige and inducible adipose tissue depot, which it responds to cold. And then we have the classical white fat, which has near undetectable levels of UCP1. And so, you know, with cold and adrenergic stimuli, we get uncoupling, and brown and beige fat utilizes these fatty acids through futile cycling rather than exporting them through lipolysis. And this is depicted here with a cold stimulus where we get signal transduction beta-adrenergic stimuli promoting a lipolytic cascade, and instead of these fatty acids being liberated, feeding the liver, they're actually being oxidized internally through this futile cycle. And why do we think this is important? Well, ultimately, you know, we published many papers over the last few years showing that activation of the brown adipose and beige adipose tissue can protect mice from developing fatty liver disease, and the system is quite sensitive to that, independent of changes in adiposity, suggesting this sink is quite powerful for protecting the liver. What we know is that under standard housing conditions that we all use for mice of 20 degrees Celsius, that the energy expenditure is much, much higher than an average free-living human. And this energy expenditure is really driven by this overactive brown and beige adipose tissue, which is sucking up all this excess substrate. And one of the ways we can minimize this activation is by housing mice at thermoneutrality or 30 degrees Celsius, which much more closely mimics the actual energy expenditure of a free-living human. And we have this hypothesis that, you know, this might be one of the reasons why we've been able to correct NASH thousands of times in a mice, but still, you know, 20 years later, don't have a drug on the market. And so, you know, the key question here is, can we develop a mouse model that better replicates the development of human NASH by housing mice at thermoneutrality? And so what Marissa did in this scenario was have mice fed a chow diet or NASH diet in this situation, sort of the classical amylin diet, high fat, high fructose, but with physiological levels of cholesterol, not super high levels of cholesterol, which we think is very important. And what she showed here is that the NASH diet certainly promoted steatosis and ballooning as expected, but when we housed mice at thermoneutrality, this effect was even greater. And this was independent of any change in body mass or glucose tolerance or insulin sensitivity, okay? These parameters were very comparable between the groups, whether they were at room temperature or thermoneutrality. We also saw increases in inflammation and NAFLD activity score over time. And this is after 24 weeks on this diet. Of course, we're all interested in fibrosis, and here we saw that the fibrosis development at thermoneutrality was much higher in the mice fed the NASH diet, as shown here using Picasaurius red, as well as trichrome staining and other sections of the data. And so pathologically, what we could see, what Marissa and our pathologist, Elham, showed was that these livers showed many of the pathological symptoms of advanced human NASH, ranging from glycogen and nuclei, malaried dank bodies, microglanulomas, lipogranulomas, as well as microvesicular and macrovesicular structures, suggesting that we were pushing the disease pathology to a more severe stage. Consistent with that hypothesis, when we did transcriptomic analysis, what we could see is that these thermoneutral NASH diet-fed mice were much more closely associated with more advanced F3, F4-type patients, suggesting that we were accelerating disease development. And this is shown here, also in this gene signature correlation matrix, where, you know, thermoneutral NASH diet is much more effective than sort of the room-temperature NASH diet in eliciting many of the characteristics of advanced NASH patients. So we think this is an important aspect of the disease pathology. And what do we think might be going on here? Well, when we keep the mice happy at a nice, warm temperature, they're not needing to put on any clothes or shiver or keep up their brown fat. What then happens is that we have this excess free fatty acids, rather than being oxidized within the beige and brown adipose tissue. Instead, this is feeding into the hepatocyte, much like it does in a human. As we know, this is the major source of acetyl-CoA under nutrient-replete conditions. We also, in the context of NASH, we have high levels of glucose and insulin driving glycolysis. This also generates acetyl-CoA. And then if, in the absence of excess demand or any increase in demand, we'll get an accumulation of citrate, which has to go somewhere to keep this TCA cycle moving. And where does it go? Well, it gets exported out of the mitochondria and is then converted back to acetyl-CoA by ATP citrate lyase. And then on to malonyl-CoA from acetyl-CoA carboxylase, and then malonyl-CoA gets converted to fatty acids through fatty acid synthase complex. So ACLY also generates oxaloacetate, which contributes to high levels of glucose through the gluconeogenic reaction, and acetyl-CoA is a key substrate for the sterile synthesis pathway. And we know collectively, you know, high levels of fatty acids, cholesterol, glucose, insulin is sort of the toxic milieu in which NASH develops. So our previous work has really highlighted the role of AMP kinase in this aspect, phosphorylating both acetyl-CoA carboxylase and HMG-CoA reductase, manipulating this pathway. And our other studies have really highlighted, you know, we can mimic many of these effects with an ACC inhibitor. But about 10 years ago, we got thinking, well, if this is so effective, what might happen? And indeed, you know, many of these things have moved on to clinical trials, APK activators, phase II clinical trials, ACC inhibitors, as many of you will be aware of, also effective in lowering steatosis, but, you know, having the downside of increasing serum triglycerides. So clearly, the risk profile could be improved. So about 10 years ago now, we hypothesized, well, maybe we could elicit similar effects by working downstream on the pathway by inhibiting ATP citrate lyase directly, and, you know, maybe having all these beneficial effects that we get with AMPK activators, but working more directly. So the data I'll show now is really a project that's been involving looking at whether inhibiting ACLY exerts therapeutic effects in NASH and NASH-driven hepatocellular carcinoma. And so, you know, the first model we used was the ACLY flocks to animals generated by Katie Whelan, which she kindly deposited at Jackson Laboratories. We used a TTR-CRE, a V8-CRE-driven system, and we could see we had very effective deletion of ACLY in the hepatocyte. As expected, we saw large reductions in sterile synthesis, fatty acid synthesis, and increases in fatty acid oxidation within these hepatocytes. If we made the mice obese, insulin-resistant, housed them at thermoneutrality for 24 weeks, as the intervention I just showed you, and then injected them with either the YFP or that CRE adenovirus, what we could see is that the ACLY knockout mice had some protection from steatosis, ballooning, but really no change in inflammation with this inhibition of ACLY in the hepatocyte only. There are quite dramatic reductions in fatty acid metabolites. Liver triglycerides went down. And importantly, there was no really compensatory increases in serum triglycerides and serum cholesterol, and we can discuss where that's going later. But we also did a study where the pharmacological inhibitor of ACLY, known as bempedoic acid, which has been approved by the FDA for hypercholesterolemia, and we found that in this intervention model using the ACLY flox mice injected with YFP, that this agent had similar effects on steatosis, ballooning, but also had this really nice effect on inflammation. And in contrast to the hepatocyte-specific ACLY knockout, also lowered fibrosis. So this was a differential effect of the drug compared to the genetic knockout. And this led us, we were quite puzzled for this, what was happening here. So of course we ran an RNA sequencing experiment and found that fatty acid metabolism was the top up-regulated pathway, DNL was up-regulated, gluconeogenesis was up-regulated, consistent with what we might anticipate if we are down-regulating these pathways. But really what jumped out at us with the drug treatment was that we had this really pronounced effects on extracellular matrix synthesis and collagen biosynthesis, which was really not observed in the knockout. Of course, everyone here knows that these pathways are tightly associated with hepatic stellate cells. And indeed, when we looked at this using RNA-scope, we found that even in the ACLY hepatocyte knockout, there was very strong staining of ACLY within alpha-SMA staining cells as well as active 2A staining cells as shown here. So this is in the ACLY hepatocyte knockout. We're using that very specific CRE driver, but there's still the ACLY activity expressed in the hepatic stellate cell. And so Ziggy then did a series of experiments where she pharmacologically activated human and mouse hepatic stellate cells and then inhibited ACLY using bimpedoic acid and found that it effectively lowered activation state of these hepatic stellate cells, secretion of procollagen 1A alpha. And in addition to suppressing this activation state, it was also lowering cell proliferation. So there's a whole body of evidence suggesting that ACLY is an inhibitor of cancer cell growth, which I'll get to next. But this was really inhibiting stellate cell proliferation activation as well as their proliferative state as shown here using FACS and the Ki-67 marker of the hepatic stellate cell. So this suggested that ACLY inhibition was having an effect in NASH. Consistent with this, we can see that ACLY expression is quite highly expressed in the NASH liver. But what is quite interesting here is that actually the highest area of ACLY expression is in NASH-driven HCC, where the ACLY expression is very, very high within the tumor. And indeed, if you look at the TCGA data sets, you can see that people with high expression of ACLY within the tumor have a, you know, a much lower survival rate compared to those with the low ACLY activity, suggesting that it may be doing something. So we wanted to investigate this, and this is work by Jaya Guantan, who's here at the meeting. And using our thermoneutral model, what she did is that using the ACLY-floxed animals, she injected them at two weeks of age with diethylnitrosamine, which induces DNA adducts and the genetic instability, switched them to the thermoneutral housing, the high-fat fructose diet, for 20 weeks, followed by the CRE injection. And what we can see here is that genetic deletion of ACLY in this context dramatically lowers the number of visible tumors within the liver. You can also see improvements in steatosis visually here. And importantly, you know, the number of tumors was dramatically reduced. Many mice had over 90 tumors in this model, whereas the ACLY knockout mice had none. And if we look at the number of neoplastic lesions, basically the ACLY knockout, many of the animals had no neoplastic lesions. So none of the most aggressive tumor types in the ACLY knockout, many of the ACLY knockout animals. If we look at what happens pharmacologically using a novel ACLY inhibitor with better tumor penetrating capabilities compared to benpedoic acid, which will be presented more on poster 236.1 on Saturday afternoon, using the same experimental paradigm, Jaya found a very similar reduction in tumor burden in a dose-dependent manner, where the I32 lowered visible tumor levels by approximately 70 percent at the highest dose. So what might be going on here? Well, if we look at the tumor histology, what we can see quite striking in both the ACLY knockout and the drug-treated tumors is this rapid and, you know, quite pronounced delipidation of the tumor. So both the knockout and the drug are not just affecting the liver, they're really lowering both fatty acids and cholesterol within the tumor. And we know that we measure this biochemically as well, where the fatty acids and cholesterol levels go down quite dramatically. And when we ran RNA sequencing on the tumors, what we could see was quite surprisingly what was happening here is this reduction in tumor lipid burden seemed to be dramatically enhancing T-cell and B-cell infiltration into the tumors of the ACLY knockout animals. And we did this at two different time points, one at a late time point and one at an early time point before there was even any difference in tumor development, suggesting, you know, this was a causal factor driving the reduction in tumor burden in the ACLY knockout animals. If we look up chronically what happened, well, we can see that the ACLY knockout and drug-treated animals had marked increases in cleave caspase 3, indicating increases in apoptosis, as well as this reduction in proliferation, suggesting that ACLY is inhibition in the tumor and the liver in hepatocytes is really not only suppressing proliferative capacity but also promoting tumor apoptosis. And so, in summary, what we've talked about here today is really this concept where targeting ACLY downstream of AMP, upstream of AMPK-ACC inhibitors not only has an effect on lowering steatosis and fibrosis independently of increasing serum triglycerides but also seems to be lowering the proliferative capacity of hepatic stellate cells and maybe also increasing tumor immunogenicity and promoting apoptotic pathways. And, you know, we don't have the answer for the exact mechanism here, but we'd like to speculate that potentially this increase in fatty acid oxidation in the tumor is increasing availability of glutamine and glucose for activated T cells to home in on the tumor and this switch to this metabolic reprogramming is key for that. So the key takeaways here are that thermoneutral housing of mice and feeding a diet high in fat and fructose replicates many of the metabolic, pathological, and transcriptional characteristics of human NASH. I don't have a fancy name for these mice like we heard before, but maybe we'll call them the hot mouse or something like that. There you go. And then we know that inhibition of ACLY increases fatty acid oxidation and lowers serum cholesterol and triglycerides and blood glucose as well as hepatic steatosis, bleeding, and fibrosis in mouse models of NASH. I've shown one such model. We used another model, the STAM model in the paper. Inhibition of ACLY reduces hepatic stellate cell activation and proliferation and also reduces HC's proliferation and promotes tumor immunogenicity and apoptosis. And so with that, I'd like to thank the funding sources and, you know, the great team that did all this work. We have Marissa, the lead author on the NASH, Bempidioic Acid Study, and then Ziggy did the stellate cell, Jin Han is all the transcriptomics, and Jaya and James are the key leaders on the liver cancer work. So thank you. We'll do the panel at the end. So I'm sorry, last but not the least, we have a very exciting talk on hepatic microRNA regulation of metabolic homeostasis by Dr. Fernandez-Hernando from Yale. I would like to thank the organizers and Natalia for the invitation. I'm a cardiovascular researcher, and I have to disagree with you that the heart is still very important. The heart requires the lipids made by the liver, but it's a very important organ. Okay, you can say that. My talk is, I'm going to talk about microRNAs and specifically about one, miRTC. This is like summarize what a microRNA is and how the mechanism here operates in the body. MicroRNAs are genes that are encoded in the genome, and they're encoded in the genome in different regions. We do have microRNAs that are regulated by their own promoters, microRNAs that are encoded in the introns of genes. In most of the cases, these microRNAs are being transcribed when the gene where it's encoded is activated. That is the case of the family of microRNAs I'm going to talk to you later. These microRNAs are exported to the theta plants through this protein called sporting phy that are further processed by another endonuclease called Dicer to generate the mature form of the microRNA that is loaded in the RNA silencing complex. The RNA silencing complex, no thanks, okay, the RNA silencing complex binds to the three-prime and translated region of the genes and controls gene expression by two different mechanisms, most of the case by promoting translational repression and also mRNA degradation. The regulation of gene expression through microRNAs and how this controls many biological processes has been complicated to address, and one of the main questions for that is because a single microRNA can't control the expression of multiple mRNA targets. They make it very difficult to identify the function. And some of these mRNA targets can be also regulated not by one microRNA but by many others. The case that I'm going to show you today is somehow very specific because this microRNA controls a specific pathway in a very specific manner and controls many components on those pathways at the same time. The discovery of this microRNA was done by my laboratory in collaboration with Dr. Catherine Moore. There were a number of papers came up after this, I think five to 10 papers reporting exactly the same thing. That was the identification of this microRNA in the intronic region of the SRAP2, S-R-A-U-F-2, sterile response element binding factor 2, which is the key transcription factor that regulates lipogenesis in mammals. The interesting thing that we found here is that this microRNA is encoded in the intron of the gene, and it's absolutely conserved across the species. As you can see here in this cartoon, within a gene the exons are usually highly conserved, but the interest don't. But in the case here, you're seeing this pink island sequence for microRNA 33 is conserved across all of them. Then what is this MRCCC doing, and I'm going to summarize in a couple of cartoons the data that we already know. As I mentioned to you before, this microRNA in mammals, there is a family of microRNA, and they encode in the SRAP2 transcription factor, and the other, miR33b, in the SRAP1 transcription factor. They actually control, together with the SRAP transcription factors, the homeostasis of cholesterol metabolism and fatty acid metabolism, and also phospholipid metabolism. This will come up publishing soon. And they do through a complementary mechanism. We do know that the steroid-responding binding protein, too, stimulates the enrichment of cholesterol in the cell by promoting lipoprotein uptake through the upregulation of the LDL receptor, as well as many of the genes involved in the cholesterol biosynthesis, including the HMH-CoA reductase. The co-transcription of miR33 target the key cholesterol efflux transporters, including the VCA1 and G1. However, the locus, this functional locus, operates in these two sense, having some major role to promote the enrichment of cholesterol in the cell. In the other case, we have SRAP1, miR33b, SRAP1 is the key transcription factor, control lipogenesis, particularly fatty acid synthesis, activating the expression of ACC, fatty acid synthesis, many of the genes that you previously heard in other talks. And miR33a also inhibits significantly the expression of the key regulators of the fatty acid oxidation, including CpT1-acro-ADHB. This is interesting because the discovery of these microRNAs also link both pathways. We know the role, the studies from Bruce Spigman, Goldstein, and Brown separate both pathways. We do know now that miR33a and miR33b can control in both ways either fatty acid metabolism as well as cholesterol metabolism. Then the story was quite elegant at this time because not only control lipid metabolism at cellular level, we do know that miR33a actually is a key regulator of organismal lipid metabolism. And this is represented in this cartoon and summarizes all this work from our laboratory where we identified that miR33a controls, for instance, this process called reverse cholesterol transport that is being regulated by the expression of this protein in the liver called ABCA1, which is a mandatory protein that is requested for the generation of biogenesis of the high-density lipoproteins. These high-density lipoproteins interact with all the cells in the periphery, including those macrophages that are accumulated in the arterial wall, where ABCA1 plays also a key role promoting cholesterol efflux. And this cholesterol is given back to the liver through the SRE1 receptor and then eliminated through the bilayer excretion. I want to point out here that miR33a is not only regulating the reverse cholesterol transport controlling ABCA1, but also regulate fatty acid oxidation through this target that has been validated by our lab and many others, including mp-tine, STPT1A, and CROT. It can also control insulin sensitivity through this receptor, IS2, and other groups have shown also that regulate bilase synthesis and bilayer secretion, therefore controlling the whole process. Then in the last, you know, few years, we tried to also understand where is the potential contribution of miR33 in the context of NAFLD and NAS. This is a recent study that's shown that both transcription factors, SREP1 and SREP2, appears to be highly upregulated in NAFLD and NAS, either in the human population as well as in models of NAFLD NAS. When they do upstream analysis of potential transcription factor that regulate the transcriptomic signature on those livers, they found that both transcription factors appears to be highly upregulated in both human and mouse studies. Then the model that we use here is the model from Matthias Heckenwalder from Germany, who fed mice with choline-deficient high-fat diet for three, six months, and after 15 months, these mice started developing hepatocellular carcinoma. This model, I'm not going to show you the data, but this is published elsewhere. These mice developed dyslipidemia, they developed insulin resistance, obesity, several of the features that are found in people with NAFLD NAS. Then in the upper panel, I show you what's happening in the stage of NAFLD, and the lower panel, the thing that happened in NAS, as you can see here, one of the things that we are able to see here is the levels of miR33 actually are increasing in both conditions. And after three months in choline-deficient high-fat diet, the same thing is shown in six months. The body weight decrease in these mice, and there is a significant improvement in the glucose tolerance, shown here in the glucose tolerance test. There's also improving in insulin sensitivity, as shown here, by the ITT assays in both in during the NAFLD stage, and NAS is also improved. In NAS stage, we also observed a similar thing, increased levels of miR33 in the liver, and also improvement in the glucose tolerance in these mice. Then this is in the NAFLD stage, then we actually did a study of the morphological study in the liver, as well as other parameters, as shown here, the liver weight is significantly reduced in the hepatocyte-specific knockout mice for miR33, that correlate to a significant reduction in liver triglycerides. This significant reduction in liver triglycerides was associated with an increase in fatty acid oxidation, shown here, also by seahorse experiments, as well. I will conclude later that this is probably likely mediated through the increased activation of MP kinase, which is a bona fide target for miR33. We not only see this increase in fatty acid oxidation, but also the fatty acid synthesis and the cholesterol biosynthesis appears to be significantly reduced in these livers. We did a morphological characterization of this liver and see a very dramatic improvement, not only a significant reduction in total T content in the liver, but also by, as shown here, a significant attenuation of the oral rate, the oral staining, also when you compare the wild-type mice with the hepatocyte-specific knockout for miR33. And this also correlates with the increased levels of MP kinase and phospho-MP kinase. We do know that MP kinase, phosphorylated iso-4-alpha for MP kinase is significantly reduced in Nafl, as shown here, and this is partially risky when you analyze the livers from the hepatocyte-specific miR33 knockout. Then this is in the states of NAS. This model generates a very strong inflammatory reaction in the liver after six months in this particular diet, as shown also previously in the Nafl DS states. We also, again, see a significant reduction in the liver size compared with when you normalize with the body weight. And also the historical analysis of this liver showed that the valunium, for instance, is significantly attenuated as well in the HNE. We further did an analysis of fibrosis in these mice. These mice have severe fibrosis, shown here in the series red, and again, it's significantly attenuated in the knockout. That correlates also with significant accumulation of neutral lipids in the liver as well. We didn't find which kind of surprising difference in the accumulation of monocyte macrophages in the livers of these mice. We validate this through different studies. This is included, for instance, here, the Westerblot analysis for fibroinectin-1 and coli, and to show again that the NAS diet induced fibrosis, as shown in the series red, but also here in the Westerblot, and this is markedly attenuated in the liver hepatocyte-specific knockout mouse model. We did also hydroxyprolinensase, and this was confirmed as well. Then what happened in these mice? These mice, as I mentioned to you, like around 50 percent of the mice, 40 to 50 percent mice develop tumors after 15 months in this particular diet, and the thing that we observe is that the mice that are deficient in MYR33 also develop significantly less tumors, and this actually correlates with the circulating levels of alpha-phytoprotein that appears to be significantly reduced in these mice compared with the wild-type mice. Then mechanistically, we have many, a lot of work here. I want only to point out this one in that the pathway, the JAF-TAF pathway, which is associated with the parathyroid carcinoma, appear to be highly affected here. It is known that the free cholesterol levels appears to be very important in the activation and stabilization of TAF that is leading to a parathyroid carcinoma, then the cholesterol levels itself are significantly reduced in the liver of these mice, and this correlates with significantly reduction of TAF levels and also nuclear translocation that correlates also with a significant reduction also in the expression of some of the mRNA target genes from TAF here in the mice that lack MYR33 in the pathocytes. Then just to finish and summarize all that I mentioned to you before, we do think that MP-Kinase is mediating part of this effect. MP-Kinase not only is itself a very important target for MYR33, and the thing that we think is happening here is that suppression of MYR33 in the pathocytes leads to an increase in fatty acid oxidation by that relating some key or bona fide targets for the MYR. Also stimulate autophagy. I didn't have time to go over that, but our group and also other groups, the group from Kathleen Moore and others, also the important effect, not only on the MP-Kinase, but also MYR33 itself in some of these targets, knocking down MYR33, at least in macrophage accelerators to stimulate autophagy. Then we think it's probably another mechanism how the whole model works, and at the same time, the accumulation of MP-Kinase, as mentioned by previous speakers, also suppress lipid and cholesterol synthesis that we think is also part of the mechanism that we see here. Thereby, suppression of MYR33 not only will attenuate enough LD2-NAS, but also a progression to HEC, at least in our system. With this, I want to finish, just only to acknowledge Pablo Fernandez-Tusi, who was the person in my lab who led most of this project and the funding agency that were instrumental to develop this project. Thank you. Thank you. We'd like to invite the speakers to come up to the podium, and we can take a few questions. Please come up to the microphones and identify yourself. So here we go. I'm Nicola Squillasche from Italy. I'm an infectious diseases specialist. I have a question for Fernandez-Hernando. What about silencing MYR33, silencing? Yeah. Yeah. We are not doing that. I mean, we had to do these experiments. We didn't do in long-term effect yet, but we're going to do probably in the future. The problem with the MYR33, I mean, also here, we have more data now. The knockout mice are obese, right, and they are obese, the global knockout, okay, are obese. The thing that we know now, for instance, is the global knockout is obese because the brain. The IGFP neurons promote the mice even more. But at the systemic level, it's fine. Then this is actually prompting us now to develop anti-MYR33 therapies to treat chronically the mice. But we didn't do the experiment yet. And you have experience of some clinical application of silencing in humans? Yes. I mean, this is a license for several medical applications, not only metabolic diseases, but also for fibrotic diseases as well and inflammatory diseases. Then it's being treated, it's being used for several diseases, but yeah, I think so, yeah. Go to this and we'll come back. Thanks very much. I'm Prakash Ramachandran from Edinburgh. It's a question for Professor Steinberg. With your hot mice, do you think this is a NASH-specific effect or do you see similar effects if you use other injury models in the liver or even in other organs? Is this a more general effect in inflammatory models? Yeah, we haven't examined that. I mean, we can comment that glucose homeostasis, insulin sensitivity was similar at this 24-week time point. We haven't done a comprehensive metabolomic profile of other organs to see what's happening, but it's something we are interested in. This work has suggested from AJ Shala that certainly the atherosclerosis can be accelerated at thermoneutrality and there's many other people who've studied thermoneutrality in other conditions and see that it does more closely mimic the human condition. Interesting. Thank you. Hi, my name is Paul Wrighton from Editas Medicine. I had a question about PMPLA-3. So what stage of the NAFLD or NASH process do you think that deletion of PMPLA-3 could help patients? Could it actually reverse or would it be more of a preventative type of treatment? And then secondly, if you do knock out PMPLA-3 in hepatocytes, what kind of side effects might you expect or to look out for? That's a lot to unpack in a short time. The short answer is it's really not known, but I can speculate. Personally, I think going early will have greater benefit. When you have established fibrosis and because it's not like there's only one pathway driving the disease, PMPLA-3 is doing whatever it does in the context of everything else that is happening. And when there are other immune inflammatory pathways that are already turned on and the fibrogenic cycle has got its own machinery turned on, just impacting one little thing, just conceptually, if I was an engineer, I would say that wouldn't be my first choice to make the intervention. That's just me. In terms of the adverse events that you might expect, if you take out PMPLA-3 altogether, is that there is a normal process of fatty acids being turned into triglycerides, stored for when you need it, and then mobilizing that fatty acid. So I'm not personally that sure that complete knockout is such a great idea of any gene, but certainly over here, it theoretically carries the potential for creating some other consequences of lipid accumulation. Thank you. Hello. I'm Chaofan Fan from German Cancer Research Center. I have two questions. The first question is for the PMPLA-3 story, because Professor, you show that the mutation of PMPLA-3 has severe NASH compared with the Y-type mice. Have you ever checked the NASH-related HCC incidence between the Y-type and the PN mutation? Yeah, that's a fantastic question. We are actually doing it right now, because, you know, in the model, the HCCs occur after 36 weeks, and so it's a long experiment. And so it is an ongoing study, because it was a lot to do this eight-week study, and then we've now done a 16-week study, and then we got a little distracted with HSD-17, and so we're back looking at PMPLA-3. So those long-term studies are ongoing. Okay, thank you. Sorry, the second question is for the ACLY story. So, Professor Steinberg, I want to ask if you find any difference about the ACLY knockout mice and the ACLY inhibitor-treated mice for the NASH severity and the NASH-HCC development. Yeah, in general, the chemical inhibitor works better. Works better. Yeah, then the knockout, and, you know, the reason we think that is, is because the chemical inhibitor also has effects on stellate cells, whereas the genetic knockout was very specific for hepatocytes. I mean, it's also possible the inhibitor has other effects, other, you know, on- and off-target effects that are beneficial. You know, we know bempidoxic acid is converted to a CoA derivative that also activates AMP kinase allosterically. So it also activates- it's a dual ACLY inhibitor and an AMPK activator. So that could explain some of the effects, but- Yes, I mean, have you ever checked the immune cells? Yeah, yeah, we did actually check the immune cells. We don't- the chemical inhibitor has no effect in immune cells because the CoA synthetase required to activate it isn't expressed. Okay, thank you. Okay, short questions, short answers, and I think we have just time for one last question. Charles DeRossi from Mount Sinai in New York. For Dr. Steinberg, have you looked in the normally housed mice and compared the two effects and what kind of differences are you seeing, if you see any? So- because most people do experiments with normally temperature mice, so I'm just wondering how it compares to when you do the thermoneutral- Yeah, so that's a good question. You know, we have done- we haven't done any experiments directly that I showed you in room temperature housing mice. I can say, you know, we've done experiments with metformin, larynglutide, exercise, all these things that have been shown to, you know, really rescue NASH at room temperature, and you know, the effects are really blunted at thermoneutrality. Metformin has no effects on NASH, despite, you know, many papers showing effects at room temperature. Exercise, no effect at thermoneutrality on NASH. So with that, we bring this session to an end. Thank you for your patience, and thanks to all the speakers. Thank you for the invitation.

fatty liver disease

NAFLD

NASH

microRNA

miR-33

PNPLA3

triglyceride metabolism

ACLY

lipogenesis

hepatic steatosis

inflammation

therapeutic targets