Hi everybody, welcome to the hepatotoxicity special interest group lecture series. Today we welcome Dr. Evangelos Triantafillou, who is a group leader and lecturer in liver immunology at the section of hepatology and gastroenterology of Imperial College London. I like that the hepatology comes before the gastroenterology. Smart. Evangelos undertook his PhD at King's College between 2013 and 2017, studying the mechanisms of immune-mediated injury in acute liver failure syndromes, and he spent a part of his doctoral work at the University of Birmingham. Okay, sorry, I got distracted with an alert on my screen. In 2017, he joined Imperial College London for his postdoctoral research position, during which he identified the role for MRTK and PD-L1 signaling pathways in regulating myeloid cell inflammatory and bactericidal functions during liver injury. The Triantafillou lab investigates the cellular, molecular, and metabolic crosstalk underlying immune dysregulation in acute, such as acetaminophen, ALF, and chronic, such as mast cell and decompensated cirrhosis, liver injury. They have a translational approach, and they use human tissue samples as well as in vivo disease models, and the focus of the lab is basically to examine dysregulated and ineffective immune responses to infection and understand how these shape hepatic immunity and influence disease progression. Another area of interest is defining the mechanisms through which different checkpoint pathways influence liver inflammation and immediate immune regulation, with the aim of exploiting these to produce new immune-directed therapeutic strategies. Evangelos is an expert in innate immunity in the liver, and we are very excited to hear his talk today on macrophages and function and heterogeneity in acute liver injury. So, Evangelos, thank you for accepting our invitation, and please take it away. Thank you so much for the kind introduction and also for the invitation. It's a pleasure to be with you online today. It's 4 p.m. in London, so good afternoon everyone, and thank you for joining online. As Lily mentioned, the focus of the talk today is on macrophage function and heterogeneity in acute liver injury, and I won't go into details about the group. I think Lily gave a very nice overview. So, we're mostly interested in immune dysregulation and immunomodulation in liver diseases, both acute and chronic, and we have been looking into innate and adaptive immunity, and more recently we have developed a research program where we focus on immune checkpoint pathways and how these regulate the cellular crosstalk and regulation in the liver, subsequently how they impact inflammation, fibrosis, and also how they alter the antimicrobial defense mechanisms. So, for today, I thought it would be good to give a broader overview about liver macrophages and then focus in the context of acetaminophen acute liver injury and what we know and how we reached that knowledge today, actually, and where we are heading in the future. So, to start with, I know this is a specialized audience, so I don't want to go into many details, but I just want to remind you that the liver is a central immunological organ. It consists of diverse cell types that perform multiple functions, including digestion, metabolism, and detoxification, and it acts as an immune organ playing crucial roles in antimicrobial defense. I would like also to remind you that it receives a dual blood supply, as you can see here, where you have portal venous blood coming from the gut that mixes with oxygenated arterial blood into the hepatic sinusoids, and this drains into the central veins. So, due to variations in oxygen, nutrients, and endotoxins within the liver, the hepatic lobules are divided into distinct functional and phenotypic areas, and we have these periportal, midlobular, and pericentral areas, and we know with a lot of studies from single-cell RNA sequencing as well, more recently, that hepatocytes present functional metabolic zonation in the liver, but importantly, also the immune system of the liver presents zonation. There's like special cell localization, and as you can see here, for example, the dendritic cells and MAID cells are more concentrated towards periportal areas of the liver, and we also have the resident macrophage population, TEMPT-Cooper cells, which resides within the hepatic sinusoids and is located between the periportal and midlobular zones. So, in homeostasis, the liver is constantly exposed to microbes and endotoxins, and therefore, it maintains a telogenic immune state in order to prevent unnecessary excessive immune activation, but the liver also has the capacity to support immune responses in response to infection or tissue damage, and also cause inflammation, such as in the case of acute liver injury, fibrosis, or a muzzle. So, what about macrophages? The liver houses an abundant population of tissue-resident macrophages, which are named Cooper cells, and which have embryonic origin and self-renewal capacity, and as I mentioned before, they reside within the hepatic sinusoids. These macrophages exceed crucial functions during homeostasis, including iron homeostasis, lipid metabolism, clearance of apoptotic or aged cells, immune tolerance, importantly, and also antimicrobial defense. So, what are the main characteristics of Cooper cells? They are sessile, they don't move, they are exclusively located intravascularly within the hepatic sinusoids, and as I mentioned before, they present with zonation. They are more concentrated in peripolar areas. So, this optimal location of Cooper cells underscores their crucial role in ensuring homeostasis by constantly clearing blood-borne pathogens, associated toxins, such as LPS and cellular debris. And for this reason, Cooper cells are equipped with high expression of FC receptors, complement receptors, scavenging receptors, or pattern recognition receptors, through which they are able to identify different death-associated molecular patterns or pathogen-associated molecular patterns. And you might be aware of some of these molecules, such as TLR9 or TLR4 and TLR3, which have been studied in the context of acute and chronic liver injury. So, through the recognition of these patterns, Cooper cells are one of the first responders in the liver to inflammatory cues, which can be sterile, such as in the case of APAP acetaminophen-induced liver injury, but also they contribute to protection from microbial dissemination and development of sepsis. If you asked me 10 years ago, and what we knew in the field, everyone would tell you that macrophages at steady state are predominantly Cooper cells, and they are marginally replaced by monocytes. But this knowledge has evolved, has changed, because recent multi-omic approaches, including single-cell RNA sequencing, special transcriptomics, and proteomics, have set new lights into hepatic immunity, and particularly in macrophage heterogeneity. So, what do we know today about macrophages in the liver? So, as I mentioned before, the majority of macrophages, around 90 to 95 percent, is residing within the hepatic sinusoids and is of embryonic origin, and they're termed Cooper cells. But these studies have revealed that in addition to the Cooper cells, we also have within the liver some non-Cooper cell populations, which have been termed accordingly bile duct lipid-associated macrophages or central vein macrophages. And in addition to the interhepatic macrophage populations, we also have the macrophages that reside within the capsule of the liver, which are termed monocyte-derived liver capsule macrophages. But things change during liver injury and inflammation. A lot of studies in the field that I have highlighted here in the bottom have examined the role also of the monocytes that are coming to the liver and give birth into monocyte-derived macrophages. So, today we have identified different subsets of these macrophages in the liver in different contexts, and we now know that monocytes can replace and take the place of Cooper cells and give rise to these monocyte Cooper cells in the liver in the context of fatty liver disease, for example, or there is an accumulation of lipid-associated macrophages termed lumps. And for instance, in the context of fibrosis and cirrhosis, a similar macrophage population, which has been named SCAR-associated macrophages, SAMS, resides within the fibrotic nodules and has been suggested to contribute to fibrogenesis. So, overall, macrophages seem to play a crucial role in cellular crosstalk and immune regulation in the liver microenvironment, and depending on the context and time point of the disease, they either contribute to injury inflammation and fibrogenesis, but equally, they are very important for resolution of inflammation and tissue repair processes. However, given this recent evidence of pathogenic macrophage populations in the liver, we require further research in order to determine their precise functions. So, what do we know about macrophages in acute liver injury? For today's purposes, I'm focusing mostly on acute liver injury and acute liver failure, which is a very rare clinical syndrome. And as you may know, hypotoxicity from acetaminophen overdose accounts for the majority of the cases of acute liver failure in North America and also in the UK. So, this is a model of sterile liver inflammation, where hepatocyte damage leads to recognition of this damage associated molecular patterns by innate immune cells in the liver, such as macrophages, and this immune sensing causes liver inflammation and subsequently can lead to systemic and hepatic immune dysregulation. And we also know that many of these patients, for example, are highly susceptible to infections, which is a leading cause of their mortality. And with regards to the clinical features, ALF, as we know, is developed with coagulopathy, jaundice, and hepatic encephalopathy, and it arises in the context of an acute hepatic injury and the absence of prior chronic liver disease. And moreover, these patients are characterized by a systemic inflammatory response syndrome and often develop extrahepatic organ failure. But our knowledge and research into macrophages has started from clinical observations and human studies, and some of our group and some from others, where we have initially identified that blood monocytes have an increased expression of CCL2 in the liver and also in the circulation, and they are reduced in the peripheral blood of patients. And equally, we know that these patients are characterized by increased levels of plasma and hepatic CCL2, which is the chemokine that binds to CCL2 receptor. And this is associated with monocytopenia and disease severity. What others and our group has shown is that ALF patients are characterized by an expansion of liver macrophages, and this is not only due to the recruitment of monocytes into the liver. As you can see here, they are characterized by an expansion of CCL2-positive cells, but also through proliferation of the resident population. And in total, there is an expansion of CD68 macrophages in the liver. What we have also shown in these patients is that further phenotypic characterization of their peripheral monocytes and interhepatic macrophages is that in the context of acetaminophen acute liver failure, monocytes and macrophages exhibit an anti-inflammatory phenotype as characterized by increased CD163 expression and reduced HLDR expression. And we have also shown that one of the molecules is particularly produced in the liver of the patients called secretory leukocyte protease inhibitor, or otherwise SLEPI, is upregulated in patients with acetaminophen overdose, is associated with infection and poor outcome, and also is able, is capable in inducing this anti-inflammatory phenotype on the monocytes and macrophages. We'll be leaving for the Carolinas on May 22nd. So, more evidence in this context of acute liver injury has come from the mouse model, which is an APAP overdose in wild-type mice. And here I'm showing you the different time points after paracetamol overdose, where the monocytes and macrophages have been extensively characterized. So, what do we know is that following acetaminophen overdose, which causes hepatocyte damage, as I mentioned earlier, pupa cells become activated and secrete a variety of chemokines and cytokines, which lead to liver recruitment of neutrophils, as you can see here, and also monocytes into the liver compartment. And this peaks in the early hours after paracetamol, between 6 and 12 hours. What we also know, though, if you can see here on the right with intravital microscopy, is that compared to a control mouse liver, which has the majority of macrophages and pupa cells, after paracetamol overdose, the liver is infiltrated with monocytes that are interacting within the sinusoids with the macrophages. And later on, these monocytes, which peak at 24 hours after paracetamol overdose, differentiate in situ into monocyte-derived macrophages. So, what do we know collectively from the studies in the field? We know that, for example, early on, the KUFA cell population is reduced by 24 hours, but it's recovered through cell renewal during the resolution phase. We also know that the monocyte recruitment into the liver is dependent on CCR2, which is the chemokine receptor that I mentioned earlier, and that this CCR2 LY6C high monocytes differentiate into LY6C low monocyte-derived macrophages that are very important for the resolution of tissue injury. And as you can see here, this CX3CR1 high population is within areas of hepatocyte damage. Furthermore, earlier studies showed that infeasible ablation of these monocytes can actually impair liver resolution. So, for example, if given an anti-CCR2 antibody, you block the infiltration of the monocytes, and therefore you block liver resolution and repair. But we also know from research from Frank Take's lab in Berlin that monocytes and macrophages have also equally important roles in earlier stages of the paracetamol overdose. For example, within the earlier time points at 6 and 12 hours, these monocytes and their descendants, the monocyte-derived macrophages, aggravate injury. And this is also shown by reduced, actually, necrosis and ALT levels when you use CCR2-deficient mice. And we also know from this work that early pharmacological monocyte inhibition to mimic the CCR2 blockade reduces liver injury and does not impair repair processes. So, collectively, this data in the field have shown that macrophages can have both tissue-destructive or tissue-reparative roles, depending on the time point of liver injury. In our group, and actually during my PhD, and this is the work that we published, I think, in 2018, we have further highlighted the importance of liver macrophages in resolution of inflammation by specifically examining the role of mert-tyrosine kinase, or otherwise called mRTK, where this receptor is the main receptor for macrophages involved in the clearance of apoptotic cell and cell debris. So, what we found is that mRTK monocytes and macrophages are expanding in the circulation and the liver of acute liver failure patients, compared to healthy controls and patients with chronic liver disease. And we also showed that these cells are characterized by a pro-reparative function. For example, they have reduced pro-inflammatory cytokines, such as TNF-alpha-9-L6, but they produce more anti-inflammatory cytokines, and they are very highly ferocytic cells. So, if you feed those cells with apoptotic neutrophils, for example, or apoptotic T-cells, they are able to clear them better. We also showed that mRTK-deficient mice exhibit persistent liver inflammation. And as you can see here, compared to the wild-type mice, mRTK-deficient animals were characterized by more necrosis at 8 and 24 hours after paracetamol overdose. And this more necrosis was paralleled by increased intrahepatic numbers of neutrophils. And from this work, we concluded that mRTK is important for the resolution of inflammation in the context of acute liver injury. Since I mentioned neutrophils, it is very important to also note that the macrophage-neutrophil interaction is also very crucial in this macrophage switch towards a prorestorative phenotype. So, this work here, this evidence supports the contribution of neutrophils into liver repair. And actually, if you look on the left side here, early antibody-mediated neutrophil depletion using an anti-LY6G antibody leads to increased liver injury characterized by increased levels of ALT or necrosis. And what this study was able to show as well is that neutrophils, and specifically reactive oxygen species produced by them are important mediators that trigger the conversion of inflammatory monocytes to pro-reparative macrophages. And here, for example, if you combine, if you block neutrophil equipment into the liver, as previously shown, you have more necrosis. However, if you reconstitute the liver microenvironment with wild-type neutrophils, you're able to reduce this damage, and this is due to the reparative macrophage population that happens in the presence of neutrophils. And these findings collectively identify a cooperation between these two innate immune subsets in orchestrating resolution of inflammation. And since we are talking about macrophages, it is also very important to note that they can also be harnessed as a therapeutic strategy in order to treat acute liver failure or acute liver injury. In this very interesting study by Stuart Fobb's group in Edinburgh, they were able to show how infusion of monocyte-derived macrophages can impact on liver disease outcome and inflammation. So for example, here they're used in wild-type animals, bone marrow-derived macrophages that were differentiated into three different types. And as you can see here, they were differentiated into classically activated macrophages in response to LPS and interferon gamma. They were also modified into alternative activated macrophages with treatment with IL-4 and IL-13, or as deactivated macrophages or immunosuppressive ones by IL-10 treatment. But what this study importantly showed is that by infusion of these three different types of macrophages in animals with paracetamol overdose, you can have different outcomes. One example, despite not changing the serum levels of these mice in ALT and AST, if you look here on this column, you can see that alternately activated macrophages were able to reduce levels of HMCB1-positive cells in the liver and also the number of neutrophils and necrosis. And also, this study showed that it increases hepatocellular proliferation. How this happens functionally is because the alternately activated macrophages are highly phagocytic in situ, and they have demonstrated that in the mouse models. And they do this also by concomitantly reducing the levels of point-flammatory cytokines in these animals. Well, for example, here you can see that compared to animals with paracetamol overdose, alternately activated macrophages reduce the levels of IL-12 or IL-6 in the serum of the mice. And similarly in the liver, for example, IL-6 is downregulated. So this suggests that maybe macrophages can be used as a therapeutic strategy in the liver after paracetamol overdose. So how about the immune response in bacterial infection? As I mentioned before, this is a major interest within our group, and you might have come across those studies in the field, but we now know that hepatic macrophage-mediated bacteria clearance is reduced in the liver after either acute or chronic liver injury. So either you give overdose of paracetamol in the mice, or if you treat them with carbon tetrachloride for a period of 6 to 10 weeks, the ability of macrophages in the liver is impaired. And therefore, if you infect paracetamol-treated mice with either bacteria negative or bacteria positive strains, you see that less bacteria are trapped in the liver and more are circulating in the blood. And this again underlines the main function of macrophages in the liver in response to bacterial infection. So as part of this study, we were trying to understand which maybe molecular targets are able in preventing this reduced capacity of macrophages in uptaking bacteria. And given that we have a broad interest in new checkpoint pathways, we started examining the role of the PD-1 axis, which is a wide immune checkpoint pathway studied in different disease contexts. For those who do not know, the PD-1 axis is a major immune checkpoint pathway and it has been studied in cancer and infection the last at least 20 years. And just to give you a very brief overview, what happens is that at steady state, the interaction of PD-1, which is expressed mainly on T-cells with its two ligands PD-L1 or PD-L2, which is expressed on antigen-presenting cells, maintains peripheral immune tolerance. However, in the context of viral infection or cancer, this axis is upregulated and contributes to immune suppression. The last couple of years, the knowledge that we have about this pathway has expanded and it has also been investigated in the context of chronic liver disease. For example, as you can see here, PD-1 axis has been studied in the context of viral hepatitis, on metabolic dysfunction associated with steatotic liver disease, and also in the context of hepatocellular carcinoma. But what we were able to show is that in addition to its central role in adaptive immune dysfunction, the PD-1 axis has a very important role in innate immune dysfunction. I don't want to go into details into this study, but this is a schematic from the paper and this is published data. So you can go and read it if you would like. So what we were able to show here is that compared to control wild-type mice at steady state after paracetamol overdose, you have an expansion as I previously described of monocyte-derived macrophages during the resolution phase and T-cells and T-regulatory cells. But these different immune cell subsets upregulate the checkpoint ligands, PD-L1, on macrophages and T-dex and KT cells, but also PD-1. And collectively, this impairs the response to bacterial infection. How does it do that? We also look as well in human external tissue and blood of patients. And as you can see, equally to the mouse studies, compared to pathological control tissue, we also saw that patients with ALF had an activated pathway in externals. So we had more PD-L1 and more PD-1 signal. What we were able to demonstrate here is using a combination of different techniques from flow cytometry and intravital imaging in mice, that if you either use PD-1 deficient mice or if you treat wild-type mice with paracetamol overdose with an anti-PD-1 monoclonal antibody, we were able actually to improve the bacteria clearance in the liver. And this is actually shown here, where, for example, compared to control mice, mice with paracetamol overdose have less bacteria capsules in the liver. And this is also reflected in the blood, where you have more bacteria in the blood. But if you pre-treat these mice with an anti-PD-1 antibody, you're able to improve the response to infection and prevent bacterial dissemination to extrahepatic organs. And this was also combined by a reduced sepsis score in mice with paracetamol overdose. Of course, PD-1 access is not important only in the context of acute liver injury, but as I mentioned before, in chronic liver disease settings. And I won't go in detail into that for today, but I just wanted to show you some of the preliminary data here as well, where we have been looking into this access in the context of fatty liver disease. And by using special proteomics, immunophenotyping, we have started exploring this access further and have demonstrated that compared with patients who have low fibrosis, patients who have higher fibrosis are characterized by an expansion of PD-1 positive macrophages in the liver and potentially more PD-1 positive CD8 T cells. So this project is founded and currently ongoing, where we are trying to understand further how the checkpoint upregulation is mediated in myeloid cells, what are the mechanisms that lead to a checkpoint upregulation in monocytes and macrophages, and how does this access regulate their function, and if it actually impacts liver inflammation and fibrosis. And finally, how it modulates the new response to infection, with the aim of finding alternative targets to modulate this pathway and prevent susceptibility to infection in patients and the animal models. So what about therapeutic immunomodulation and future directions in the context of acute liver injury? So overall, in the field of acute and chronic liver failure, macrophages have been considered as a target. However, a lot of those molecular targets or approaches have failed in the clinical setting. So various groups have targeted macrophages at different parts. So you can either prevent their activation, or you, for example, can prevent their recruitment into the liver. So some of the promising strategies, in my opinion, is the prevention of their activation, and this could be achieved through FMT in the context of acute liver failure, or as I saw you earlier, maybe through macrophage-targeted therapies. And this is also a part of our clinical study, which is ongoing, where we have been exploring how PD-1 can prevent infection in patients, but we don't have the results in our hands yet. And of course, in addition to these approaches, we might have other targets that we could utilize in order to modulate macrophages in the context of acute liver injury. So it could be potentially other immune checkpoints, such as the TIN3 or the TIGIT axis. It could be maybe TLR receptors, which are able to recognize different bacteria or damage in the liver. Or it could be different agonies and antagonies, for example, through bile acids. So how can we move forward in the field and how we can achieve maybe therapeutic immunomodulation? As I mentioned before, a lot of single-cell and special approaches have been published in the recent years in the field of liver disease. And one point that I would like to make is that mouse models are interesting, but do not always recapitulate fully the clinical setting or maybe the clinical syndrome and progression of patients. So I think what we should be doing is to combine different approaches, and that involves both preclinical in vivo studies, but also in vitro models, such as 2D or 3D co-cultural systems or spheroids. And of course, it's very important to use human tissue samples, both blood or liver tissue extracts and biopsies, in order to confirm our findings. So the use of high-throughput technologies is also very important. And for example, what has been used lately is the single-cell and special multiomics involving both transcriptomics and proteomics in the liver. And why this is important? This is important because it can provide deeper insights into hepatic immunity, both innate and adaptive. It can unravel maybe pathogenic disease mechanisms, and it can decipher better the microbial immune interactions. And for that, I can give you an example here, or maybe two examples, where two very interesting studies were published demonstrating, for example, that after paracetamol overdose, the microbiome is very important in modulating the response to acute liver injury. And this is through NYC-dependent programs and toll-like receptor recognition. And it's not only the macrophages that contribute to the pathogenesis of the disease, but also other immune and non-immune cell subsets, such as endothelial cells and activated stellate cells. And also, this is another interesting study, where at different time points after paracetamol overdose, they used single-cell RNA sequencing in order to zoom in into the parenchymal and non-parenchymal compartments in the liver after APAP overdose. And here, for example, by looking into the macrophage subsets, in accordance with previous studies in the field, you can see that compared to controlled mice, where the majority of the myeloid cells in the liver are Kufr cells, following paracetamol overdose, you have a monocyte signature that is upregulated in addition to monocyte-derived macrophages that are very crucial into the resolution phase of the disease. And what these single-cell RNA sequencing and special approaches can give us is more information about the cellular interactions between these subsets, and also how the transcriptional programs of these monocytes or monocyte-derived macrophages alter during the disease progression, and how similar they are compared to Kufr cells or macrophages. And the approaches mentioned here, in combination with the in vitro models and the human tissue samples, can lead potentially into the identification of new targets in order to develop immunomodulatory therapeutic approaches in the context of paracetamol overdose, but not only. So with this in mind, I would like to finish here and thank all of my colleagues across the department and my team, and also a lot of our collaborators within the UK and outside the UK, and I'm very happy to discuss and receive any questions. That was fantastic. It generates a lot of interest. You can stop sharing if you want, so we see you on the full screen. I think a lot of us, you know, this relates to a lot of our research in different contexts. We all do different things, but, you know, I'm a hepatocyte biologist, as you know, but macrophages play a role in all diseases, and so I'm going to start with Dr. Gao from NIAAA. Yeah, I have a couple of questions. First of all, it's a very nice talk and great work. You know, we follow your work very well. I have a question about this bacterial phagocytosis macrophage, you know, in the APAP model was significantly reduced. Do you see the neutrophil phagocytosis as bacteria or there are? From your slides, you know, you don't, maybe bacteria are just quickly killed by the neutrophils? So, at 30 states, previous work, not only from us, but others, has also shown that the Cooper cells at the healthy liver are the predominant phagocytes, and either bacteria or if you use pH draw-draw-based assays. And then maybe some of the monocyte-derived macrophages can have an important role into that, and third, the neutrophils. As part of the JCI paper, actually, we did not include this data, but I had also done inflow cytometry to see if other subsets, in addition to Cooper cells, uptake any of these bacteria early on after infection, and it was actually the majority of bacteria uptaken by the Cooper cells, like 90% of them. So, neutrophils were not really highly positive for any bacteria. Having said that, I think also the location of the Cooper cells, as I mentioned before, is very important because they are within the sinusoid. So, when you give an intravenous infection, a systemic infection, they will be the first cells to recognize and maybe trap within the sinusoid the bacteria. Well, do you see circulating neutrophils, phagocytosis bacteria? In the mouse models that we have done, I don't think I have seen. Very low percentage. Oh, well, that's very interesting. Okay, another question, because we use the coronary-induced immune-mediated liver injury model, we study how this necrotic liver lesion resolved. So, we find the APAP model and the coronary model, the necrotic lesion, I guess the resolution, the pattern is very different. So, in addition to APAP model, have you tested other liver injury models? How this necrotic lesion was resolved, eventually disappeared if the mice survived? With regard to the resolution of inflammation, we haven't really gone back to this in this context. So, as part of the Medicaid proresultive paper, for example, we have been only looking into paracetamol overdose. But we have also kind of used carbon tetrachloride single-dose overdose, but it has actually slightly different dynamics, immune dynamics, or like time points compared to the APAP. So, I cannot really make direct comparisons, I would say. I know the paper that you mentioned about your JCI work, I think you refer to the necrotic lesions actually, which is very interesting. But I think from another very recent paper, I think which was published in Science about macrophages and infection in the liver, what is very important is, as they say, you are what you eat. So, it depends if you are eating a hepatocyte, which is necrotic or an apoptotic, or it could be also apoptotic neutrophils. So, when macrophages uptake different cargo, they may switch different transcriptional programs. So, they may acquire different types as well. So, I think the model matters and also what they uptake and where they uptake, it matters. For example, it could be lipids in the context of fatty liver disease, which could change them differently. Okay. Maybe last very quick short question. Have you compared the KUFA cells and the infiltrated macrophages, phagocytosis, bacteria, or dead cells? KUFA cells, infiltrated macrophages, phagocytosis? You're not talking about the bacteria, you're talking about phagocytosis? Yeah, bacteria or dead cells as well. We have done some experiments with monocyte-derived macrophages, not with KUFA cells, but from the recent papers from the group of profs, Charlotte Scott, I would say, I think in the context of APAP as well, it is shown that not the KUFA cells present, but more of these monocyte-derived macrophages are within the areas of necrosis, where they call them lipid-associated macrophages. And this kind makes sense that, you know, they're more motile and they can move into the central vein in order to update. If that answers your question, but yeah, we have not looked into it in terms of other molecular pathways. Okay, thank you so much. Great work. Excellent. Guru? Hi, I think you have almost answered the question. I was asking similar questions on neutrophils. Is there a big difference in animal models and human paracetamol overdose explants in terms of neutrophil infiltration? Because we're not taught that that is an important inflammatory cells in this kind of injury, while there's lots of animal models seem to show it, or you just had a previous question about it. So you didn't really see it, and you showed the last slide had a very tiny group of cells where neutrophils have arrived. It depends on the time point of the mouse model, yes. But since you're trying to compare the human and the mouse context, I would say that it's very difficult to make direct comparisons with the human patient samples, because it's not very homogeneous. There's great homogeneity. Sorry, there is no great homogeneity in the patients. So some patients with paracetamol overdose may come earlier after the paracetamol overdose, some may come later on. So it's not always consistent in terms of the resolution of phase when they're coming into the clinic. So whereas in the mouse models, you can be more precise and, you know, doing phenotyping at specific time points. So maybe there's like a difference between basic tissue explants and mouse tissue explants. With regards to the neutrophils, I think in the field we have not seen really much of a defect in phagocytosis in patients, for example, on the neutrophils. But I have not examined into the mouse model into detail to give you more answer into that. Okay. Thank you. Fantastic. I think I'm next. I'm going to go have two quick questions. One is, as I bring it back always to the hepatocyte, there's, you know, hepatocytes also express PD-L1. And especially during injury, it's been shown like in being chronic injury, they can express it. Do they express it with an acute injury? Like if after I see an immunofren, have you seen any expression of PD-L1 on the hepatocytes? Because of course the immune cells expose it. Actually, we did not look into PD-L1 expression in the mouse model in the PD-L1 paper. From what I remember from the human tissue explants, we did not really see much of PD-L1 expression. In the muscle patients? In the APAP patients. Most of the macrophages, which is similar to what we saw in the mouse model. Having said that, apart to the viral infection context in the liver, there was this paper, I think, in the Journal of Hepatology, where it shows that during endotoxemia, hepatocytes upregulate PD-L1 as well. So in the LPS, you can upregulate that. So I would not exclude that as a factor in terms of, you know, in the context of... APAP may be too acute, you know? Yeah. Yeah. You may need like a more longstanding... Anyways, I just was curious about that. We call it sterile liver inflammation, but I don't think really it's sterile, because it's a lot of microbiome effects in APAP overdose as well. Yeah. Anyway, yeah. Yeah. I think actually, just in general, your talk was great. And I think the direction is very important, because as a clinician, these patients will die of infection. So it's really important to understand why they're dying of infection, you know? And most of them, a lot of them, even with acute alcohol-associated hepatitis, they're dying of infection. So we need to understand what happens to the, you know, what makes you so susceptible to infection. So that's actually very important. And then my last question was, you talked about the macrophages, and I know there's disagreement on this, right? I know I see Hartmut's here, about macrophages having a pro-injurious role in the early phase and having a more reparative role in the late phase. And my question to you was, because you have fate tracing and you have these beautiful models where you're, you know, you're imaging the cells, can you tell if they're the same macrophages? Is it an issue of plasticity? Or do these reparative macrophages leave? I mean, the injurious ones leave and reparative ones come in? Or is it the same cell? That's a very interesting point. So macrophage endogeny is not part of my research, but I'm very interested about it. I suspect, and from what I have seen in abstracts and conferences, and I think that's why I concluded with the single cell RNA sequencing and special transcriptomics approaches, that is important, at least in the context of APAP as well, to use fate mapping models, as you said, and these multi-omic approaches to see if we can identify small pathogenic subsets in the context of APAP liver injury. The previous studies in the field using CXCR1, I think, and Chimera models have shown that in the context of APAP, poop cells do not become, for example, replenished by monocytes. They are packed by 96 hours. I've been wondering about the monocytes that are coming in. Yes, in the context of APAP liver disease, it's true, and it has been shown by the group of Martin Williams and Professor Salo Scott, but it has not been shown yet in APAP, and I suspect it will be shown soon. Okay. Very, very interesting. Brian, Brian Koppel. Welcome, Brian. You don't usually join us. Excited to see you here. Sorry. I should join in more. So, great talk. So, my question somewhat relates to the transition of mouse to human in evaluating the data and potential targets of therapy. You had a very nice slide that kind of outlined all the different approaches that are being considered in acetaminophen-induced acute liver failure. One issue, though, is that almost all of those studies have been done in the context of acute liver injury and under conditions that don't produce liver failure in the mouse, and so I'm wondering how similar would those be, how translational are those to the human when it's two different things that are being studied? As I highlighted, I'm also a fan of comparing both human and mouse always. You can get some insights from the mouse models, but you cannot be absolute about it. I think with regard to the macrophage biology, there is a lot of consistency, at least in the field, in the liver field, but I think where you're coming from is more of the clinical syndrome and what mice do not develop or develop, for example, which is a similar problem in the context of chronic liver injury and cirrhosis, because mice don't typically develop cirrhosis. We use a lot of different other models to mimic this context, for example. The other approach that we haven't done ourselves, but I think people do it these days, is the use of humanized mice, but that will not give you more detail or similarities to the clinical syndrome. It will be more with regards to the immunology, maybe. So, and then also a second question that's somewhat related to your colleagues at King's College. They had shown in patients with acute liver failure that there was an association between high levels of IL-6 and IL-10 and a poor outcome in those patients. Well, it's well known that high levels of both of those, if you block them in the mouse model where you give a low dose of acetaminophen that doesn't produce liver failure, it actually makes things worse. So, clearly there's differences between mice where you investigate acute liver injury versus acute liver failure. And so, we've actually found that if you give a higher dose that induces liver failure, if you block IL-6 in those mice, you actually almost completely knock down PD-L1 in the liver and also in the recruited monocytes in the liver. So, I think there's very big differences. I mean, it depends on the question you want to ask. It's time. And for example, in our animal license, we are allowed to give higher doses of placetamol as well, because we also know that necrosis or tissue damage is proportional to the dose of placetamol that you give. So, if you give 300 milligram or you give 600 milligram, you cause different damage in the liver as well. So, I don't know if, for example, in the higher dose of placetamol, the more hepatocyte damage triggers a different program in macrophages or immune compartment that is not the same as in the early low dose, for example. So, actually, the amount of injury is exactly the same between those two doses. It's just that one doesn't get repaired and one does. And so, you know, you look at patients, if you look at the level of blood enzymes like ALT, it really is poorly predictive of outcome, because you really don't keep damaging the liver all the way out until 100% damage. You're only damaged where the P450s are being expressed for the exceeded benefit. But I think from some of our data, if I remember well, that when we compare low and high doses, there was a higher confection depletion with a higher dose of placetamol compared to the lower dose. So, that would also impact the immune disease progression or inflammation accordingly. So, it's not only the hepatocytes and, you know, the metabolism of placetamol, it's also what kind of sterile inflammatory response you trigger secondary to the hepatocyte damage. Very nice work. Thanks. Thank you. Excellent. Actually, I want to make a comment that I was thinking about this when Ben was asking his question about ConA. It's just, I'm just saying it as a comment, is that it's going to be hard to compare them because you can't control very well for the amount of injury you cause. So, you're dealing with two different models that have two different pathways and mechanisms of causing injury. And also, the severity of the injury you induce is going to impact your repair process and the way, you know, that appears and the speed of that and the kinetics of that. So, it's going to be very hard to compare two different models. You see what I'm saying? Yeah. And some models are like this are mediated immunity and some are like more innate. Yeah, of course. It's like, yeah, it's like obviously you're activating, you know, with ConA. And then, Adyan, go ahead. Dr. Apta, are you here? Oh, there you are. That's a great, great talk. I enjoyed talking to you earlier. We met last year at the ALF symposium. My question is at the molecular level, what do you think these cells do? Do they phagocytose? Do they secrete stuff? And if one of the functions is more important than the other, clearly they are pro-regenerative in my mind, at least. But what does that mean? I mean, you know, in the end, you have to fix the hepatocytes, right, to get the liver to function. How do they aid the hepatocytes at molecular level? Some of at least our data have shown that, for example, when we investigated the role of MRTK in the liver inflammation, we showed that the MRTK high subset of macrophages produces more anti-inflammatory mediators. And one of them was AGF, hepatocyte growth factor. So what these cells produce may impact in a paracline manner, maybe cell proliferation, regeneration, or other processes. Now, this is for one subset. What happens from other immune cell subsets, and they have a contradicting or counteracting role, I cannot tell you. It could be because everything is about balance, I think. What I wonder is when you, when people deplete these cells or mess around with them, has anybody gone in and checked what happens to the hepatocytes actually? It depends on the method of deletion as well, because for more than 10 to 20 years, people are using clodronate liposomes to deplete macrophages before induction of liver injury. And we now know that these clodronate liposomes are not appropriate, or potentially they have other side effects, or they can affect neutrophil biology as well. So I think moving more to genetic specific cell subs and deletion methods, it could be a bit more, you know, accurate to do. Like in the case of the mice, you have CLEK4F, for example, to do coupheset depletion. Some people don't believe CLEK4F that much as a marker, which is fine. But I think all of these multi-omic approaches will give us more information and maybe more tools in the future to specifically target immune cell populations. Or I think a lot of people are now looking like CRISPR in vivo, knock down of genes and expression. So, and the technology is moving so fast, I'm just trying to find the time to read about it and keep track actually, you know. Great talk. Thank you. Thank you. Any more questions? I'm not sure if in the chat is there any question. I didn't check actually. No, these are all from the chat. I've been calling people. That's how we do it, Evangelos, just because then I don't know who to call in which order. We've gone through all the questions, it sounds like. Oh, I see Mark is... Yeah, hi. Just a step back and a question about biliary repair and the effect there were, you know, there's a question, can these restorative macrophages, as Lily says, are evolving from, you know, single cells that dynamically change or are a separate population in the portal area that you mentioned, do they have an impact, those restorative cells on biliary repair and also anti-fibrogenic effects by secreting metalloproteases, etc.? In other words, you know, just through cytokine release, etc. Yes, I think these two subsets that I mentioned in addition to the KUF results in the steady-state liver are very recently identified, right? So the bile duct macrophages or the central vein macrophages. I think there was a paper, depends on how people describe them, if they describe them lipid-associated macrophages, but there was a paper in Cell Reports that showed that these bile duct macrophages are actually MMP12 positive. So they sell like transceptional programs similar to the lipid-associated macrophages. Now, what's their impact specifically to bile duct repair? I'm not aware, at least up to date, if it has been investigated. But I think from everyone's perspective and research questions, I think this will be answered. Right, and again, the reason why I'm asking you this question, and we're out of time, but some of these KUF cell and macrophage proliferation programs rely on the CSF1 receptor, right? The CSF1 receptor drives them. So, and there's an experience now with pharmacology to block CSF1 receptor with some drugs. And the phenotype of liver injury that's seen is not only hepatocellular or hepatocyte targeted, but also biliary cell targeted as well. So I was kind of curious what that connection was. Yeah, I see your point actually here, but I don't have the answer to this, I'm afraid. Okay, thank you. Thanks. Ben, you wanted to say something? Yeah, no, maybe I have a comment for Mark's question. We published a paper in 2021. When we induce acute bile duct damage, you know, three hours is a ton of macrophage monocyte accumulates there and repair there, repair the bile duct injury. Within three hours? Yeah, no, no, we do see, I don't want to take our time, but you can see three hours ton of monocyte accumulate around this damaged bile duct. So if you depleted them, you know, bile duct damage repair was delayed. I thought maybe that's repair the bile duct. Yeah. And Evangelos, before I let you go, we've been trying to, you know, in my immune checkpoint inhibitor work, because we do see also a macrophage infiltrate, we've been trying to do some immunostaining on them and I can't get CLEK4 to work. Have you ever gotten CLEK4 to work, the antibody on histochemistry in mouse? Yeah, we use team four, CLEK4F and F480. Actually, we're using a conjugated antibody. I can send you the details if you wish to. Can I ask you, because I'm trying to characterize these to see whether these are like Cooper cells or infiltrating macrophages. So I'm trying to kind of phenotype them a little bit, just crudely in the lab with immunostochemistry before I spend all my money on the, you know, on the big stuff. So if you have a protocol or if you tell me what antibody you use. Yeah, yeah, no problem. I think the field is moving towards more to the team four VCIC combination of markers with regards to the macrophages, because there is evidence that potentially CLEK4F or like other surface phenotype markers may be lost. So they cannot really identify reliably the resident embryonic Cooper cells and the monocyte derived Cooper cells, whereas VCIC and team four work better in this discrimination. I can send you the protocol and the details. That's not a problem. I would love that. Thank you so much. Yeah, no problem. I have read the paper, actually. It's very interesting. Thank you. And it's interesting because after we, and I know I talked to Lucia, she doesn't see this, but after we depleted the T cells, we still got a macrophage infiltration. And she said, you guys don't see that. We see lower, but we still see some. Yeah. So that's what we see too. So, but I'm trying to see what's going on there and what is, you know, what are those, what are those macrophages, you know? So that would be helpful to like phenotype them. Yeah, no problem. I can drop you an email. Thank you so much. Bye guys. Thank you for the invitation again. And I might be coming to us in November. I haven't sent an abstract yet, but we might see each other there if you are going to the... Oh, I'm definitely going to be there. I would love to see you. Yeah. I will let you know. I will keep you posted. Yes, definitely. I'll be there the whole time. Thank you again for the invitation and everyone for coming online. Bye Evangelos. Bye-bye.

hepatotoxicity

macrophage function

acute liver injury

immune dysregulation

acetaminophen-induced liver injury

Kupffer cells

MRTK signaling

immune checkpoints

therapeutic strategies

single-cell RNA sequencing