Hi, everyone. I am Lily Dara. It is my pleasure, along with Dr. Jim Leyendyk and the rest of the Hepatotoxicity SIG Steering Committee at the ASLD to introduce to you our speaker today, Dr. Brian Kopel from the Department of Pharmacology and Toxicology at Michigan State University. His talk is titled, Transitions from Acute Liver Injury to Acute Liver Failure, Lessons from Animal Studies. Dr. Kopel is very well-known in this field. He has a PhD in Pharmacology and Cancer Biology from the University of Nebraska Medical Center and the Epley Institute for Cancer Research. He did a postdoc with Robert Roth in the Department of Pharmacology and Toxicology at Michigan State and then spent eight years at Kansas Medical Center. From there, he returned to MSU and has been there since. His research largely focuses on elucidating mechanisms of immune cell regulation in liver disease. He's had many seminal publications, including he's known for his work on hypoxia-inducible transcription factors, his work on PDGF-beta and how they're produced by macrophages and other liver cell types during fibrosis development. He's also demonstrated a role for bilases in triggering hepatic inflammation during cholestasis, which is something I'm very interested in. And currently, his studies aim to uncover the mechanisms of ALF and how we transition from acute liver injury to acute liver failure and what is the tipping point. And he has some very interesting data and some very interesting theories about the role of the vasculature in that regard. So take it away, Brian. Thank you for joining us. We're very excited to have you. Thanks. Thanks for the invitation. So this is actually kind of a new area for me. My, as Lily said, my background's mainly in regulation of immune cells in the liver and we focused a lot on bile acids and also hypoxia and how that affects those cells. However, here, I'm gonna talk more about acute liver injury and acute liver failure. And I'll bring this up a little bit later, but I have Uday and Apti to actually blame for me getting into this area. So, yeah, I'm gonna talk about transition from acute liver injury to acute liver failure. I was going to talk about the vascular component to this, but as I was putting this together, I just, I kind of realized it was just, it was too much. So what I'm gonna talk about more of today is really some work we've done, but also a lot of work that other people have done in this area and try to come up with a, an idea of how to study this, especially in mouse models. So, as you all probably are well aware, there are a lot of causes of acute liver failure. So viral hepatitis, autoimmune hepatitis, there's various metabolic and genetic disorders that can lead to acute liver failure, but also oftentimes the cause is indeterminate. By far, the greatest cause of acute liver failure are drugs and especially acetaminophen inducing drug induced liver injury. So in most instances, probably greater than 95% of the time, when the liver is injured, it triggers a reparative response that then rapidly restores liver function to the proliferation of hepatocytes and other cell types in the liver. However, under certain conditions, patients can actually go from this acute liver injury and transition to acute liver failure. And when this occurs, these pathways that allow for the restoration of liver function become disrupted by mechanisms that aren't fully understood. So this is fairly common in that liver injury is repaired and goes back to normal. And it's fairly rare instances where acute liver injury transitions to acute liver failure. And we need to learn more about this because this is really associated with a high mortality rate, which actually a long time ago used to be very high, but due to improvements in critical care medicine, mortality rate for this is improving. So what is acute liver failure? So I have to admit, so I'm not a clinician, I'm a basic scientist. So when I started working in this area, I actually really wasn't aware of what the definition of acute liver failure was. I always thought that it was essentially just the liver stops functioning normally, but really clinically it's defined as acute liver injury that occurs in the absence of any other evidence of liver injury or disease. It's usually shortened duration and it's associated with changes to coagulation. So you have to have the development of a coagulopathy, but also you get hepatic encephalopathy. And this is a neural disorder occurring in the brain. And this encephalopathy actually can lead to cerebral edema and increase intracranial hypertension, which can be lethal. And actually this is a terrible disease to have. I currently have an aunt who has very severe hepatic encephalopathy from MASH. So there are other things that happen, however, and that includes multi-organ failure. So you can have renal injury, you can have lung injury and other types of injury. You can have hemodynamic alterations, so hypotension. So a lot of these patients require pressor agents. You get a lot of metabolic disruption also. And one of these that occurs that may contribute to the encephalopathy is an elevation of ammonia. So the liver's ability to clear ammonia becomes compromised. And then as I mentioned, the coagulopathy. So another thing that occurs in a lot of these patients is they develop a condition called systemic inflammatory response syndrome. And essentially what this is, it's a case in which severe liver injury causes a spillover of various pro-inflammatory cytokines into the systemic circulation. And then it's thought that these actually can contribute then to multi-organ failure. And also I'll provide a little bit of evidence for this, that these cytokines in combination with ammonia may actually contribute to the encephalopathy that occurs. So just like with any inflammatory response, you get a trigger of a anti-inflammatory response in order to turn off the inflammatory system. And this also occurs in these patients. And oftentimes these patients have very high levels of IL-10 and TGF-beta. And I think a few months ago, you had a talk on PD-1 and PD-L1 in acute liver failure patients and in animal models. And this produces a condition called compensatory anti-inflammatory response syndrome, which is a state in which the patient is very immune suppressed. And obviously this is detrimental from the standpoint of these patients can develop infections, which could lead to sepsis and contribute to death. All right, so what I became interested in is why under certain circumstances, does the liver not repair and it progresses to acute liver failure? And I think the consensus from a lot of clinical studies is that the severity of the hepatocyte injury is really not the determinant of this process. And so studies that have looked at survival versus non-survival in patients have found that they essentially have very similar levels of hepatocyte injury. So why is it important to elucidate this? If we can understand how this occurs, we could potentially identify biomarkers such that you could identify a patient that's going to continue to deteriorate so that they can be transferred more quickly to a liver transplant center to be assessed for a liver transplant. And also this can allow us to identify therapeutic interventions to actually prevent this from occurring. And can we restore this process in all patients? The first thing we asked is, is there actually an experimental approach to investigate this? And it's important to have an experimental model system because this is very rare in patients. So the patient population is very small. And so it's really hard to look at this in a mechanistic way. And so we looked at whether there is an experimental approach to investigate this. So that led us to this paper here by Udayan Apti. And so how I became interested in this brought his graduate student at the time, who's now at the University of Pittsburgh, was actually presenting this at a meeting. And there were a couple of things from this study that kind of really sparked my interest. And in fact, I actually told myself when I finished my postdoctoral work that I would never treat a mouse with acetaminophen. And part of the reason for that is, and I just looked the other day, currently since the early 1960s, there's about 10,000 publications on acetaminophen and liver. And so I kind of figured that where could I actually even make an inroad here and actually come up with something interesting. And also there was a lot of controversy in the area with a number of different studies having conflicting findings. And so I really wanted to avoid this area until I saw Udayan and Bharat's work here. And so what they found in these studies is that if you treat mice with a dose of acetaminophen of 300 milligram per kilogram, you get hepatocellular necrosis in the liver. And this is primarily within the central ovular regions of the liver where the cytochrome P450s are located that metabolize the acetaminophen. So this produces severe liver injury, which over time then begins to resolve. So monocytes are recruited, they clear out this dead cell debris and ultimately hepatocytes proliferate and restore the liver. And this works very well. The liver is very good at doing this. However, if you increase that dose to 600 milligrams per kilogram, one thing that interested me about this is that the amount of liver injury is actually fairly similar between these two doses. And there is a little bit of an increase here, but for the most part, they're pretty similar. And in our hands, these two doses fairly consistently produce the same amount of necrosis. And so what they found is that this higher dose though, what happens is the necrosis occurs, but then it persists in the liver. And there is a failure in the mechanisms to clear this dead cell debris. And that's where I became interested in this is what role at this higher dose or how are monocytes impaired in this process of clearing the dead cells is disrupted at this higher dose. And could that be a contributing factor to the progression to ALF? So essentially, I became curious as to whether comparing these two doses, could we better understand what's driving acute liver injury to acute liver failure? So similar to patients, there's a similar severity of liver injury. However, in one instance, this injury persists, the parasites fail to proliferate here and the mice ultimately die. So essentially, could this represent patients that develop acute liver injury? So for instance, this would be essentially a patient who develops severe necrosis, but they fully recover essentially without any intervention. So these mice, they don't receive an acetylcysteine, which is the antidote for acetaminophen overdose. They haven't received presser agents or anything else. They recover just fine. However, at this higher dose, could this represent acute liver failure where you fail to actually repair the liver and ultimately leading to liver failure? And also another reason I became interested in this is could this actually explain some of the controversies in the field? So in other words, could these pathological differences explain numerous conflicting findings in the literature? And one of these that has come up numerous times is neutrophils. So with neutrophils and with a lot of other cell types and mediators, you can find papers where it is involved in the injury and you can also find papers where it's not involved in the injury. And so I became curious as to whether, is this just related to a dose difference? And in fact, if you actually start looking at the literature and comparing these conflicting findings, oftentimes the dose is different between the two studies and that's really the only difference. So the pathological differences between these might explain those conflicting findings. And recently Hartmut Jeschke's lab kind of confirmed this by showing that if you deplete neutrophils at this low dose, it doesn't impact hepatocyte injury at all. However, if you deplete neutrophils at this higher dose, it does cause a modest reduction in the liver injury. And so really this could potentially explain why there's so many conflicts in this field leading to all these editorials that you see all the time. And so I actually tried to get a student of mine during COVID to write a review on this, to kind of go to the literature, find all these conflicting studies and see whether it really kind of turns out to be that it's dose differences. And unfortunately she was wanting to move on to a job so that never got done. But another example of this is toll-like receptor 9. So at this higher dose, if you treat mice with toll-like receptor 9 knockout mice with acetaminophen, it protects them from liver injury. However, if you block TLR9 at this low dose, it actually makes things worse. And so TLR9 is protective at this low dose, but then is actually promoting the injury at the higher dose. And so I think the point I'm hoping to make here is that it's very important to consider the dose of acetaminophen that you're using when you do these studies. All right, so more than half of all AL cases are the result of acetaminophen overdose. So one thing that's at least nice about this model is it's clinically relevant. However, I would say a majority of the studies of the 10,000 studies that are out there investigating acetaminophen-induced ALF are actually using doses of 300 milligram per kilogram or less. And so really this is really not investigating ALF, we would argue. This is actually investigating patients that would essentially recover on their own. And so one thing I'm not saying is that these studies aren't important. And one thing I would say about the field of acetaminophen is that much of what we know about how the liver repairs itself after injury really has come from these studies here. And it's really applicable to other tissues after injury. So a lot of important information has been discovered from using these lower doses of acetaminophen. So essentially what have we learned from these studies? And I was hoping to, most of this work that I'm gonna summarize here actually is from other labs. I was hoping to put in the references but it just got to be too many. So unfortunately I don't have all of that in here. But acetaminophen is metabolized by cytochrome P450s to this toxic metabolite in acetyl P-benzoquinone amine when high doses of acetaminophen are taken. This NAPQI then produces hepatocyte injury by mechanisms that have been reviewed numerous times. So I'm not going to talk about that here. Presumably then damaged associated molecular patterns are released from the hepatocytes and these damps can stimulate then Kupfer cells in the liver to produce pro-inflammatory cytokines. And so for example, some of the cytokines that are produced would be interleukin-6 or TNF-alpha. And the damps seem to stimulate production of these cytokines through release of HMGB1, DNA, other things like ATP released from dead cells activating pattern recognition receptors like the RAGE receptor and tolike receptor 4. In addition to these cytokines, the Kupfer cells can make these neutrophil chemotractic factors and they can also produce CCL2 which is a chemotactic factor for monocytes. And when that occurs, oh, sorry. Another thing that we have actually worked on and also Jim Leyendeik's lab has worked on is we have been investigating through a collaboration with Greg Lemke at the Salk Institute, investigating the role of Axl in acetaminophen-induced liver injury. And what we found is that Axl is activated by a protein in the blood called GAS6 when it's bound to phosphatidylserine. And so since these cells are injured and they're dying and the membranes are breaking open, that would expose phosphatidylserine which then can activate GAS6 or bind to GAS6 which activates Axl. And what we found from these studies is that this process then stimulates upregulation of MMP12, a matrix metalloproteinase, which then Jim Leyendeik's lab had shown to be protective to the endothelium in the liver. This was what I was going to talk about was more of this, but I thought it was important to kind of talk about, talk about the model more than focusing in on this. So if you're interested in the vascular aspects of this, you can kind of look at our work with Axl and Jim's work with MMP12. All right, so the neutrophil chemokines... Oh, one other point to make is that one thing that is a bit of a mystery with acetaminophen-induced liver injury is that after the damage and after these Kupfer cells release cytokines, they essentially disappear within the lesions. And I don't think it's clear what happens to these cells. It's possible they're dying by pyroptosis or some other type of mechanism like that that then leads to them being lost within this region. But numerous studies have shown that that is the case. So these chemokines recruit neutrophils into the liver. CCL2 recruits monocytes through the CCR2 receptor. These monocytes, when they accumulate in the liver, they're considered to be pro-inflammatory, so they do express some inflammatory cytokines. And they are marked by having high levels of LY6C and high levels of CCR2, which is the chemotactic receptor for CCL2. And they express low levels of F480. So oftentimes then it's proposed that the neutrophils then eventually die by apoptosis. And the monocytes recognize this and phagocytose these neutrophils that have died and also phagocytose dead cell, excuse me, dead cell debris that's within here. When that happens, it actually stimulates these monocytes to mature into macrophages. And in doing so, they lose expression of LY6C and also CCR2, and they gain expression of F480. Sorry, I'm developing a cold. My whole house is sick here, so I'm also developing that. All right. Whoops. So when this happens, these cells produce high levels of IL-10 and other anti-inflammatory mediators, which then are involved in resolving the inflammation. And then work through other groups, including Udine, APTI's group. IL-6 and TNF-alpha are released and can help to stimulate hepatocytes to proliferate. Also, Wnts are released from sinusoidal endothelial cells and contribute to this. And what this does is it ultimately causes the hepatocytes to proliferate and to restore the hepatic parenchyma. In addition, the Kufr cells proliferate in the liver to restore Kufr cell numbers. So this is what happens when you essentially treat a mouse with 300 milligram per kilogram of acetaminophen. And this is not everything that happens. So there's a lot of work out there that I didn't talk about. Platelets and fibrin accumulate in here. Also, there's a role for NK and NKT cells, lymphocytes in this process. And unfortunately, I just didn't have time to include all of them. Sorry, I have a visitor who's a sick kid from, who's home sick, sorry. So we wanted to follow up on these studies from Udine and look at whether these mice actually develop clinical features of liver failure. So in other words, do they develop hepatic So in other words, do they develop hepatic encephalopathy and do they develop evidence of coagulopathy? And so what we did first is we treated mice with these, a low dose to 300 milligram per kilogram of acetaminophen and the higher dose. So we've used 500 and we've also used 600 milligrams per kilogram. For some studies, we've used 500 in order to lower, to make it less lethal to the mice. And so what we did is we looked at cerebral blood flow. So there's been a number of studies that have looked at cerebral blood flow in patients with acute liver failure. And especially in patients with drug-induced acute liver failure, oftentimes cerebral blood flow is reduced in the brain. And so we looked at this in collaboration with Anne Dorrance in our department and used laser speckle imaging to image blood flow in the brain. And so in this imaging, red staining is high blood flow and blue is a much lower blood flow. And as you can see, there's little difference between vehicle treated mice and mice receiving 300 milligrams per kilogram. But what we found in the higher dose mice is that there's a reduction in cerebral blood flow. And this occurred in multiple regions of the brain, including the frontal lobe and parietal regions. And so this was consistent with findings from patients with ALF. So next we looked at neurological endpoints in these mice and we used the guidelines set by the International Society for Hepatic Encephalopathy. And we looked at various reflexes in these mice to see if there's evidence for hepatic encephalopathy. And so essentially how this is, is there's various measurements. A score of two means that it's normal. A score of zero means it's completely absent. And then a score of one is kind of an in-between. And you can see with respect to these various measurements, there's very little difference between vehicle and 300 milligram per kilogram mice. However, when you get to this higher dose, you can see that there's a fairly substantial reduction in these various neurological endpoints. So one thing that also occurs in patients is that you get leakiness within the blood brain barrier. This can lead to cerebral edema, which can ultimately lead to cerebral hypertension. And so we looked at this by measuring the water content of the brain and found that at 300 milligram per kilogram, it's really not different between that and vehicle. However, there's an increase in the water content in mice treated with the higher dose, indicating that they are developing cerebral edema. So one of the contributing factors that's been proposed to the development of hepatic encephalopathy is blood ammonia, and that ammonia builds up in the blood, it accumulates in the brain, and it can have various effects leading to the encephalopathy. So we also looked at ammonia, and this is blood ammonia concentrations. And what was surprising is that really at the low dose, the ammonia concentration is elevated to the same extent as the high dose treated mice, suggesting that ammonia alone is probably not sufficient for the hepatic encephalopathy in the mice. And then it may require other factors that are present at these higher doses. And we have some evidence that it might be involved inflammatory cytokines. And this is consistent with some patient studies where you can see elevations in ammonia in patients. However, they don't display symptoms of hepatic encephalopathy. So it really is kind of suggested that ammonia in combination with something else is probably contributing to this. So what we concluded from this is that mice treated with the 500 milligrams or 600 milligrams per kilogram develop features of hepatic encephalopathy that are consistent with what's observed in patients with liver failure. So the next question is, is there evidence for coagulopathy? And Jim Leyendeich's lab actually looked at this. And so they looked at various measures of coagulation in the blood. And as you can see, most all of these show some, sorry, evidence of coagulopathy, especially at the higher doses in the mice as opposed to the lower doses. So really consistent with these mice developing coagulopathy. So essentially, we concluded from this that at doses of 500 and above, these mice most likely display, or they do display features of acute liver failure. Whereas at the lower dose, this really is consistent more with acute liver injury that repairs. All right. So the next thing we asked is, is there evidence for these other conditions that develop in the patients that have the worst outcomes? And that's systemic inflammatory response syndrome, and also CARS or compensatory anti-inflammatory response syndrome. The reason we looked at these is that there has been a number of studies, whoops, ah, there's been a number of studies showing that, sorry, I have a really sensitive mouse here. There's a number of studies showing that high levels of IL-6 and also IL-10 are associated with the worst outcome in ALF patients. And so this is one of these studies here showing that in patients that survive, the levels of IL-6 and IL-10 are low. However, in those patients that died, there's higher levels of IL-6 and IL-10. And these patients clearly had a worse outcome compared to those that had low levels. And there's been a number of studies that have really confirmed these findings. And recently a study by Hartmut Jeschke confirmed these findings. And what was nice about this study is they looked at these over time. And so, as you can see in patients that did poorly or didn't survive, there's high levels of IL-6 and IL-10 early, after development of acute liver failure. And also CCL-2 levels remain elevated over time. So we looked at these same pathways in mice treated with the different doses of acetaminophen. And what we found is, and this is looking at liver mRNA levels, that levels of IL-6 are much higher in mice that received the high dose of acetaminophen. Similarly, we see higher levels of IL-10 in mice that received this high dose. And then we also saw this elevation and persistence of CCL-2. And so one thing that amazed me is how well these lined up with each other, comparing the mice to the patients. And I actually didn't even realize this until I was putting this talk together. And so it was kind of a nice surprise to see that. So we also confirmed in the blood for elevations of these proteins also. So it's not just that there is elevations in the mRNA in the liver. Okay. So what we decided to look at next then is although there's evidence for IL-6 and IL-10 elevations in patients with the worst outcome, is this essentially just a consequence of having a more severe disease, or is it actually causally involved in this process? And so if you recall, in mice treated with the high dose of acetaminophen, these areas of necrosis persist in the liver. And so this suggested that potentially the process of recruiting monocytes or the monocytes' ability to clear this dead cell debris is somehow disrupted in mice with the high dose. So one thing we asked is, since IL-10 is an anti-inflammatory cytokine and can likely prevent recruitment of monocytes into the liver, is the elevation in IL-10 responsible for this defect in the clearance of dead cells? So we first looked at accumulation of monocytes in the liver. The marker we used here was for CD68, so it's also present on Cooper cells. So this is not just the monocytes using this marker. We have some studies starting with LY6C trying to confirm these findings, but the antibodies for LY6C are just not as good. So in mice that received the low dose by 24 hours, the arrows are pointing to central veins. And so you can see here these areas of injury that are around the central vein, and the CD68 positive cells are stained red. And you can see that they start to accumulate in the lesions. And then by 48 hours, these lesions are just largely filled with these monocytes and macrophages, which then continues to shrink until ultimately it will go away and the dead cells will be cleared. However, when we gave the high dose, you can see here where the central veins are, and you can see that these areas are completely absent of any of these CD68 positive cells. And when we quantify this both by looking at the immunofluorescence, but also by flow cytometry, there's a substantial reduction in the numbers of these cells that accumulate into the liver. And so it appears that there's a disruption in this process. So then we determine whether this is the result of high levels of IL-10. And so to look at this, we treated mice with a high dose of acetaminophen. We then treated with isotype control antibody or an IL-10 neutralizing antibody, beginning at 24 hours after the acetaminophen. And we did that because there's studies showing that IL-10 levels earlier are protective for the hepatocyte injury, and we wanted to avoid that. So as you can see here in the isotype control mice, there's very few CD68 positive cells in these areas. But when we block IL-10, you can see it somewhat restores the accumulation of these cells into these lesions. So does that restore some of the removal of the dead cell debris? So if you look at isotype control treated mice, and you look at these lesions, you can see there's very few inflammatory cells within these lesions. When we blocked IL-10, however, you can see that they start to accumulate within these areas, and this resulted in a reduction in the necrotic area. And so we treated these mice with the IL-10 antibody beginning at 24 hours, and that's where peak necrosis had occurred. And so this reduction is not due to blocking liver injury. It's actually most likely due to increased clearance of dead cells. So at the same time we were doing this study, Jim Leyendeich's lab on the other side of our building was doing a study where they took mice treated with the low dose of acetaminophen, and you can see CD68 positive cells in the lesions. And then they actually gave them recombinant IL-10. So they kind of did the opposite study of what we had done. And when you do that, you can see there's fewer macrophages, monocytes accumulating in these lesions. And when that occurred, there was actually a reduction in the number of CD68 positive cells, and actually it resulted in an increase in the amount of necrotic cell debris. And so from these studies, what it suggests is that exaggerated levels of IL-10 impair monocyte-dependent repair functions. We don't know if this is due to a reduced accumulation of the monocytes. I think if you remember from earlier on, I showed data for CCL2, which is the chemokine that recruits them to the liver, and it's actually at much higher levels in both patients and in the mice with acute liver failure. So it's not due to a defect in CCL2 release from the liver, but we're still exploring the mechanism. So another part of these studies looking at clinical ALF patients, they found that high levels of IL-6 actually correlated with a higher incidence of encephalopathy. So suggesting that potentially IL-6 could contribute to hepatic encephalopathy. And so we looked at this by treating mice with 600 milligram per kilogram acetaminophen. Two hours after, we gave 250 micrograms of IgG control or IL-6 neutralizing antibody. And we looked at cerebral blood flow. And as you can see here, after acetaminophen, there's a reduction in cerebral blood flow. And when we blocked IL-6, it seemed to restore that to some extent. And that is quantified here, as you can see, and this is total brain blood flow. Also, when we looked at these various neurological assessments, these different measurements, we saw that with the IgG, we had a low level on these different neurological score, but it was increased by giving the new IL-6 neutralizing antibody. So at this time, we had let some of these mice go longer to see if they would actually fully recover and survive. And unfortunately, they all ultimately, even though they seem better here, they all ultimately get worse off and die anyway. So what we decided to try was, well, what if we increase the dose of the anti-IL-6 antibody? Maybe we just don't have levels long enough to maintain this. And so when we did that, we actually found the opposite results. So this is survival in the mice. All of these mice were treated with acetaminophen. We gave them 250 micrograms of the antibody. There's a fairly high survival in both groups at 24 hours. However, when we increased the dose of the antibody, nearly all of the mice died that were treated with the anti-IL-6 antibody. So clinically, this is definitely not an approach to treat ALF patients. And so kind of what we have concluded from this is that although IL-6 at very high levels can be pathologic, you do need to have some IL-6 on board in order to help maintain the mice. And there's a lot of evidence that IL-6 is protective in mice that are given the low dose of acetaminophen. So to summarize this, IL-6 and IL-10 levels are highest in ALF patients with the poorest prognosis. These cytokines are protective at low dose of acetaminophen, and that's from other studies. However, exaggerated levels of these, IL-10 seems to disrupt liver repair, whereas IL-6 worsens hepatic encephalopathy. However, I wouldn't consider either of these probably clinically viable options of blocking. They all have their problems. So the next question we ask is what is the cellular source of these cytokines? Because maybe from that, we can get some insight into restoring them to normal levels and normal regulation. So to do this, we first looked at IL-10, and we used IL-10 reporter mice. So these mice, the GFP expressions driven by the IL-10 promoter, we treated with acetaminophen, we perfused and isolated the non-parenchymal cells, and then looked at flow cytometry. And I'm only showing the F480 data here, so that would be Kupfer cells. And this is at 24 hours, so it's most likely not monocytes that contribute to this. And so what you can see here is that in mice given the high dose of acetaminophen, IL-10 levels are much higher in the Kupfer cells in these mice. We did look at a lot of other cell types, and there is an increase also in NK and NKT cells. However, the proportion of those cells relative to the Kupfer cells is much, much smaller. So their contribution to the IL-10 production is probably not that great overall. So we also took these mice, and we isolated non-parenchymal cells, and used magnetic beads to purify the Kupfer cells and the monocytes. And we confirmed that in mice treated with a high dose of acetaminophen, IL-10 and IL-6 mRNA levels are highly elevated in the Kupfer cells, but we don't see this effect in the monocytes. So we wanted to confirm these findings, and what we did next is we confirmed this through analysis of single-cell RNA sequencing data. So this group here had published recently a study where they did single-cell RNA sequencing on mice that were treated with vehicle 300 milligram per kilogram or 750 milligram per kilogram of acetaminophen. At three hours, they isolated non-parenchymal cells and analyzed by single-cell RNA sequencing. And we took this data set, and we reanalyzed it using TrailMaker, which is now part of Parse Biosciences. So, and we had the software, this is the UMAP, and we had software identify the various cell types, and it identified a number of sites that were Kupfer cells. We wanted to more specifically confirm we were looking at Kupfer cells, so we looked for the cells expressing CLK4F, which is a marker of Kupfer cells, and you can see it ended up being this small part here. We took these Kupfer cells, and then we further subdivided it into different clusters here. And then we looked at the different treatment groups to see if there were any changes in these different clusters. And essentially, what we found is that at the high dose of acetaminophen, there was an expansion of this cluster three and a decrease in cluster number one. So, what we were interested in is what are these expressing? And so, we took this cluster and looked at the genes that were expressed by them. And interestingly, the highest expressed genes in this cell cluster were IL10, IL6, also both CXCL1 and CXCL2, and then PDL1, which has been implicated in immune suppression and acetaminophenidase ALF. So that was really interesting. So what we wanted to do next, then, is further look into this data set to see, can we find some mechanistic insight into this? And what we found is that we found there's an increase in this atypical I-kappa-B protein called NF-kappa-B inhibitor zeta. And actually, we had noticed this from some data that Dr. Aptey's lab shared with us, where they had done gene array. And they also, in that gene array, this is higher in the livers of mice treated with a high dose. So NF-kappa-B inhibitor zeta is a known regulator of IL6 and IL10 in psoriasis. And I think there's been some recent studies implicating it also in MASH. So we wanted to see, is this involved in amplifying these cytokines? So we first confirmed that this is actually elevated in mice treated with the high dose. And you can see there's a substantial amount of protein for this in the liver. And then we used edekonate, which is an inhibitor of NF-kappa-B inhibitor zeta. And we took mice that received the high dose. We then treated them with this two hours later. And we found that there was a reduction in IL6 and IL10, suggesting that NF-kappa-B inhibitor zeta is a regulator of these cytokines. Just one note, this edekonate, it's a fairly dirty way to inhibit NF-kappa-B inhibitor zeta. So we certainly need to confirm these findings using knockout mice. So just to summarize overall, after acute liver injury, there's an accumulation of neutrophils, monocytes in the liver. The monocyte spagocytose dead cell debris, they change their phenotype, and they resolve the inflammation. Pro-regenerative factors are produced that then stimulate the hepatocytes to proliferate. However, at this higher dose of acetaminophen that produces features of acute liver failure, there's much higher levels of IL10 and IL6, and also of these neutrophil and monocyte chemokines. The neutrophils accumulate at higher numbers in the high dose, and they actually contribute to the liver injury. Monocytes, some of them do accumulate in the liver, but they seem to stop right at the edge of these lesions, and they no longer will move into these lesions to clear dead cell debris. And the high levels of IL6 systemically may interact with ammonia to worsen hepatic encephalopathy. And so what we've been curious in, then, is essentially how are these mechanisms being disrupted to actually lead to acute liver failure? And we have some ideas behind that, which actually deals with the vasculature, but unfortunately, I didn't have time to go into that, maybe later on. So I got to thank the members of the lab who did all of this work. And then also at Michigan State University, Jim's lab, we've collaborated for I don't know how many years, a lot. And also Cheryl Rockwell, my wife, who did our flow cytometry and was nice enough to let us use her instrument, and Dorrance for help with the laser speckle imaging. And then also with Diane and Bharat for sharing gene array data that kind of helped just to get to some of this. And obviously, the NIH for the funding. Hopefully, NIH remains. It still gives funding, we'll see. So yeah, I'm happy to answer any questions. Sorry. Thanks, Brian. That was great. We have plenty of, Lily has a comment that appears to be three paragraphs long, so I'm going to cycle back to that one in a little bit and let you address each of those topics. But I think we'll start with a question from Johnny Sexton. Do you want to unmute and ask? Sure, yeah. So I was just curious if you have seen the recent spatial transcriptomic study, mainly focused on the signaling in the hepatocytes. So it's really interesting this transition from survival to death at different doses. And your description of the immune-mediated signaling here is really beautiful. It's just fantastic. So my question is, do you know how this dovetails into the hepatocyte signaling? They showed that MAPK, Nrf2, and PPAR signaling, as well as some other functional outcomes like oxidative phosphorylation capacity and induction of stress-related pathways. How does this immune signaling dovetail in with the hepatocyte repair reprogramming program? And do you know if any of these specific pathways are the most differentially regulated due to this pro-inflammatory signaling that's happening? No, so we haven't looked at that. I've really avoided looking at the hepatocyte injury just because there's so much work on that. But it is possible. So one thing that we think is that it's possible that the Kupfer cells are being activated differently because of what's getting released from the hepatocyte. So if you just look histologically at the hepatocytes between the different doses, they look very different, the injury. And so I'm curious as to whether things such as mitochondrial products and other things may get released at the higher dose, which are then activating different pattern recognition receptors on the Kupfer cells. So causing much higher levels of IL-6 and IL-10 and also activating NF-kappa B inhibitors zeta. But I mean, we're just kind of getting into that. Getting Kupfer cells from a liver to actually do these types of studies is not trivial. So it takes a lot of mice. But yeah, so I think that's a possibility. We haven't looked specifically, though, at pathways in hepatocytes. And that would be interesting to do. Great, thanks. Dr. LeMasters, do you want to unmute to ask your question? If not, I've got it covered for you. Wait just a second. Okay, I'll unmute. And so my question is that patients with alcoholic liver disease, which is a chronic condition, sometimes develop acute alcoholic hepatitis, which is something different and is often fatal. And so are the same mechanisms you describe involved? And I just add that Luca Ruth added to that question because the same things, certain things should happen in viral hepatitis, autoimmune hepatitis, and so forth. So I wonder if you could comment on that sort of outside the acetaminophen arena. Yeah, so we haven't looked at other models. It's certainly possible the mechanisms are different. So outcome for the acetaminophen overdose patients is much better than with some of these other causes of acute liver failure. So it could be different. As far as the kind of a more chronic, or acute on chronic liver failure, like with alcohol. Yeah, we haven't looked at that either. It would be great if somebody could get a model that really was working around that to where you could get disease and then have the kind of a hepatitis that can develop and look at some of this. And so, I mean, the long answer is, or the short answer is that we haven't looked at that and there certainly could be differences. We don't know. Cool, and I think another question from Dr. Ruth in the chat that I think is similar, and that is focused on like viral hepatitis, autoimmune, same type of answer. You think in those conditions just needs to be looked at? Yeah, so, yeah, it needs to be looked at as far as models for those in mice. Not aware as good of models to study those. So we haven't really looked at that. We can certainly move on to those other types of injury to see if this is similar. Trying to think if there was evidence from the clinical studies that showed similar increases in IL-10 and IL-6 and other causes of acute liver failure. And I can't recall off the top of my head whether that was the case. The complexity there, I think, is also the fact that the adaptive immune response, it's not a clean model. There's also a huge contribution of like B cells, T cells. And so it's just really hard to model those, right? We don't even know what the antigen is in AIH. And when they go into liver failure, it's not a pure like innate immune response or you know what I mean, which isn't the case in the CDC. Yeah, yeah, yeah, yeah. So isolation, right? Lily, there's another, I'm sorry. There's one question in the chat that I want to cover. Do you want to ask one and then I can move on to the chat? No, no, finish the chat first. Sorry, attempting to manage. Dr. Shiraka, do you want to unmute yourself to ask? I'm sure. I'm not doctor yet, but yeah. No, you can be, it's fine. On Zoom, everybody gets a free interview. Even better, even better. I had a question. So do you think the initial spatial arrangement of the tissue damage is actually dose dependent and might contribute to the development of ALS? So, well, with acetaminophen, I think the injury is really limited to where the NAPQI is being formed. So the P450s are mainly expressed in the central lobular area. And they're the ones, they're the enzymes responsible for producing the NAPQI. And so the injury is really often limited to there and doesn't move outside of that. And that's, I think, why the injury is so similar between the two doses, even though the doses are very different. So now I'm trying to remember your question. Sorry, can you ask that again? I lost my train of thought. Yeah, so I was wondering if even in a localized area, if the spatial arrangement of the initial tissue damage, so for example, if it's really highly clustered, it could lead to faster development of ALF because it's harder to clear areas of high clustered cells or something of that nature. Yeah, so one thing I was going to talk about and I decided not to try to talk about it was that there is one clear difference between the two doses that's not a hepatocyte injury. But in the high dose, there's clearly a lot of congestion within the injury and hemorrhage. So there's something more going on within the lesions of the high dose mice that's not occurring in the lower dose. And it could be that the endothelial cells are being damaged. And we have some evidence for that. And the pathway that I mentioned with AXL going to MMP12, that pathway doesn't seem to be present at the high dose. And we can actually, there's a reduction in gas six, which activates that AXL. And if we give back and restore gas six, we can actually reduce some of that congestion and hemorrhage occurring. So I think what's happening is probably at a high dose of acetaminophen, NAPQI is at such high concentrations, it's getting out, it's killing other cells in addition to the hepatocytes. And then that's kind of contributing to that pathology. But that's just a hypothesis. So apparently Dr. Apte has to run to a meeting. So he's going to ask a question real quick and then bug out. Brian, that's great talk. Thank you so much for acknowledging our work as well. I have a quick question about using this concept for maybe more diagnostic purposes. And it's getting clearer that with dose dependence, same molecules kind of have different contexts, right? I think there's also data, I don't think it's published, but Bharat might be able to talk about that more. AGF has probably similar kind of effects. So is there a way to develop sort of an algorithm or something where at, depending on the serum levels of these cytokines, you could kind of gauge the transition from acute liver injury to acute liver failure. And that helps the clinicians kind of get a feel within the first two days after hospital admission, whether this person's going to move on to failure or not. Yeah, I mean, ideally that would be great. I know with the IL-10 and IL-6, most of those were, it was looking within populations of ALF patients, ones that survive, ones that die. So there's not a lot of studies actually looking at ALI and ALF. There was one from the ALF study group where they looked at ALI patients. And actually some of those patients do die, even though they don't develop hepatic encephalopathy. But they didn't look at any of these other measures. There's been a number of proteomic studies or metabolomic studies, but oftentimes they compare acetaminophen to vehicle. It's the wrong comparison. I think what you mean is comparing ALF to ALI and see what's different. So we don't have the money to do that. So I think it would be great if somebody could go do that. It would be, maybe it would provide something, some information. Thank you. I got to hop off, bye guys. Yeah, thanks for getting me involved in this. I didn't want to, but. Yeah, Lily, you want to get your questions? Yeah, I have so many questions. I think we can talk for hours about a lot of this. So just a few big ones, and then I'll get to my comment. I really wanted to make my comment just because I thought there may be people on this call wanting to study acetaminophen. And I kind of want to put in like, you know, Anup and Hartmut's paper, the guidelines paper in the chat to say, please follow this. I'm tired of reviewing papers and referencing that. Please do. So there's a lot of. Yeah, I agree. There's a lot of acetaminophen, you know what I mean? So I won't, I'll leave that for last. I'll just, what I wanted to ask you is that with the, have you or actually Jim looked at, I guess Jim looked at the overexpression. If you, does it affect the outcome if you upregulate IL-10 or if you increase, I guess you dose with IL-10. I understand that the injury repair is delayed and the macrophages don't come in to clean up the debris, but does that affect the outcome? Do the mice then die? Yeah, so we haven't taken it out that far. We haven't, or Jim's lab was doing those studies and they didn't go beyond 48 hours. The opposite. So if you inhibit, can you, if you inhibit, do they survive? So I, yeah, I failed to mention that. So yeah, no, they don't. So even though you start to restore some of the clearance, what happens is the hepatocytes don't proliferate. And so some of the things that Udayan looked at where the hepatocytes, their proliferation kind of fails at that high dose, that's still happening even if we modulate IL-10 or IL-6. So yeah, there's something. So, and what it could be, I think what it could be is that at that higher dose, the endothelial cells are dying. I think that impacts how the Kupfer cells get activated, but it also prevents release of Wnts, which are coming from those cells. So Udayan showed if you replace, if you activate those pathways back and also that it can actually restore the hepatocyte proliferation. But what's interesting is those mice also don't do well. And the necrosis persists. So I've kind of wondered if you target both, if you give back Wnts and block IL-10, now will they survive and go on? And I don't know the answer to that. Yeah, it's very interesting because I know Udayan's gone now, but we often have this dilemma clinically speaking. And in general, I'm sure you're aware that, when you have an idiosyncratic DILI liver failure, the mortality is a lot higher than acetaminophen liver failure. And I've always had this theory that it's probably like failure of regeneration because the insult may be a little bit, it's not like an acute hit, but I don't know, I don't have an answer to that either. My question number two to you had to do with the sinusoids. And I wanted to just tell you that it will be very informative. And I know you didn't cover it today, but do you remember, Laurie Delev has a paper with Neil from 1997, but I sometimes cite that when I write stuff, but they use, I actually pulled it up, they use 750 milligrams of acetaminophen, but they use Swiss Webster mice. So I think what would be informative is to understand, if you have a way to isolate L-sex, is to understand if the dose dependency, she clearly showed that they actually die of acetaminophen. So they must either be in vitro. So if in C57 blacks, right? That is what you're using. I think J is what you use, right? You use the Jackson mice. If there's a dose dependent effect on the L-sex death. Yes, there is. So at the higher dose there's evidence of the endothelial cells dying just from, so we've been doing some immunofluorescent staining of those cells just to see if they're still intact and still there after that injury. And it does look like there is some injury to those. And I think also, I don't know if Jim's lab did this, if you look at hyaluronic acid levels. So hyaluronic acid is cleared by the endothelial cells in the liver, that the levels of hyaluronic acid are higher at the high dose, which would at least indicate their functions impaired. And then in the single cell RNAC study that I referenced, they looked at the sinusoidal endothelial cells by electron microscopy. And at the high dose, they have lost their fenestrations by three hours as compared to the 300, which still retained some of those. So there's certainly differences in what's happening to the sinusoidal endothelial cells. So, and we've been interested in trying to figure that out. A lot of times we actually see the hemorrhage, right? And to me, you know, I used to think that that's just collapse of hepatocytes and just, but you know, obviously Laurie has shown that there is actually toxicity to the endothelial cells themselves. Anyways, I think it's really fascinating, Brian. This was an amazing talk. And I think that that direction you're taking with looking at the vasculature would be very helpful for patients understanding that. Yeah, and I'll agree with one of your earlier comments. I'm, I get, I get a little tired of reviewing papers that are studying P450 inhibitors. That's how it is. Awesome. All right, well, we're at close to 10 after. I will go ahead and volunteer Dr. Koppel to answer any and all emails that are sent to him with follow-up questions on the talk. And I want to thank everybody for attending. Thanks for taking the time. And Brian, that was great. Thank you very much. Fantastic talk, Brian. Yeah, thanks a lot. All right, bye everybody.

acute liver failure

acetaminophen

animal models

liver injury

cytokines

IL-10

IL-6

hepatic encephalopathy

necrosis

immune responses