Good morning, everyone. Thank you for joining us for being at the NIH for this talk today by Dr. Cynthia Ju. She is currently the professor and vice chair for research in the Department of Anesthesiology, McGovern Medical School, University of Texas Health Sciences Center at Houston UT Health. She has a PhD in drug metabolism and toxicology from the University of Toronto in Canada, followed by postdoctoral training at the NIH at NHLBI. She joined the School of Pharmacy at the University of Colorado as an assistant professor and was then promoted to full professor. And subsequently, she was recruited to UT Health and joined the Department of Anesthesiology. Her research has focused on understanding cellular and molecular mechanisms of acute and chronic liver injury. She's very well known in the acetaminophen field, and she's also delved into ischemia reperfusion, as we were just talking about. Her specific expertise is in innate immune cells, non-branch immune cells in the liver and their role in tissue injury and repair and in models of both acute and chronic liver disease. She particularly has a lot of work on macrophages, neutrophils, and currently very focused on eosinophils as cells that promote tissue repair. We are very interested in hearing this talk, which is titled Eosinophils in Drug-Induced Liver Injury. Thank you for accepting ASLD's invitation. Cynthia, take it away. Thank you, Lily, for the kind introduction. I'm very happy to be here and honored to be invited to talk about this topic in this forum. I'm going to give a very brief introduction. Eosinophils, for those of you who are not terribly familiar with these cells, they are a type of granulocyte derived from the bone marrow, the same lineage as the basophils and mast cells. What we learned from textbook in terms of the functions of these cells are two major points. One is they're cytotoxic cells, they kill parasites, and they're very important in host defense against parasite infection. The other bystander effect is they cause allergic, they participate to allergic reactions such as asthma. However, this textbook description of these cells have been significantly updated by more event of genetic tools and antibodies. Actually, some of these functions that used to be associated with eosinophils were not right. These cells are currently thought to be immune modulators. This is a review about 10 years ago already. You can see here that the eosinophils interact with many different types of cells in the immune system. They can present antigens and prime T cell activation. They can also guide what type of T cell response to have in certain situations because they release factors to promote Th2 response. They prime B cell production of IgM. They release growth factors for plasma cell to survive. They are important in maturation of dendritic cells. They trigger mast cells to release histamine. They educate macrophages and they regulate neutrophils. Quite a significant new understanding of eosinophils. In terms of eosinophils in drug-induced liver injury, there have not been a whole lot of studies. There are two clinical reports. One is from hepatology. There's a study done by Dillon Network. They show that based on histology, you can see eosinophil accumulation in the liver in situations of drug-induced liver injury. In this earlier study, I think it's a meta-analysis, what they found was patients with eosinophilia and accumulation of eosinophils in liver have better prognosis or patients with severe drug-induced liver injury. If they have eosinophils in the liver and blood, they have a better prognosis in terms of survival. In terms of a preclinical mechanistic studies, there have been two reports. One was done in this model of halothin-induced liver injury. They show that eosinophils throughout for production promotes or contribute to halothin-induced liver injury. The second study at the bottom, this study focused on liver regeneration after partial heptectomy, but because carbon tet injury-induced liver regeneration is similar to partial heptectomy, they also use that model. They showed that eosinophils throughout for place protective role or mostly in promoting liver regeneration after carbon tet. We stumbled upon these results because we were, as I mentioned earlier, we were studying the mechanisms of liver ischemia reperfusion injury, and we found patients with orthotopic liver transplantation have a rapid recruitment of eosinophils in their liver. So we not translate, but we use the mouse models to study the mechanism. And what we found was summarized in this cartoon here, that when there's liver damage, there is a cytokine called IL-33 acting as an alarmin. It binds to the IL-33 receptor called ST2 on eosinophils. And we found that the most predominant, almost exclusively, that eosinophils express the IL-33 receptor, not other cells. And this signaling pathway leads to IL-13 release from eosinophils, which suppresses neutrophil recruitment and activation, thereby protecting the liver against ischemia reperfusion injury. So after this study, we wondered if this protective function of eosinophils is specific to only the ischemia reperfusion injury, or it's a general phenomenon in other type of liver injury as well. So we asked this question that, are eosinophils involved in a ciminophil-induced liver injury? If yes, how? And second question was, how are eosinophils recruited in the liver after injury? So these questions were addressed by two talented postdocs, Dr. Long Xu and Dr. Yang Yang. So the first thing we did was we measured eosinophils in the liver from patients with overdose Tylenol-induced liver failure. And this is immunohistochemical staining to detect EPS, eosinophil peroxidase, which is a specific enzyme only expressed in eosinophils. As you can see here, in normal healthy livers, we can detect maybe a couple per field of eosinophils, but the number significantly increased in patients with a ciminophil overdose. And when we took it to mice, we can do a time course. As shown here, the quantification of eosinophils in liver, we only measured at eight hours. It could be earlier, but you can see that the number goes up significantly from eight to 48, peaks at 48. So now that we see the eosinophils also accumulate in the liver after a ciminophil overdose, we wanted to see if they have any function. So the first experiment we did was to deplete these cells using anti-IL-5 antibody. IL-5 is an important growth factor to promote eosinophil development, but it's also important for any chemokine to work in terms of eosinophil recruitment. So with anti-IL-5 antibody, you will not get eosinophil accumulation or recruitment into the liver. And what we saw was anti-IL-5 mediated eosinophil depletion resulted in an increase of ALT levels, shown in the gray bars here. At eight hours, it's not significant statistically, but 24 hours, there's a dramatic increase of ALT and larger areas of liver necrosis. Can you see my arrow? Yes, we can see it. Okay, thank you. And to confirm this observation, we did another way where we give wild-type mice bone marrow derived eosinophils immediately before a ciminophil challenge. And you can see here the mice received eosinophils have a significant reduction of ALT levels and liver necrotic areas. To further corroborate this finding that perhaps eosinophils play a protective role, we compared wild-type with eosinophil-deficient mice. These are double GALA-1 knockout mice. GALA-1 is a transcription factor that's important for eosinophil development. So these mice don't have eosinophils. So we had three groups. These open circles are wild-type. The open squares are the eosinophil-deficient GALA-1 knockout mice. And the closed circles are those knockout mice adoptedly transferred with wild-type bone marrow eosinophils right before a ciminophil treatment. So you can see the knockout mice had increased elevation of ALT levels at 24 hours. But if they were given eosinophils, the ALT levels decreased significantly at both 8 and 24 hours. And this is in agreement with the histology evaluation in terms of areas of necrosis. As I mentioned, we were initially wondering if this is a general phenomenon that eosinophils play a protective role in acute liver injury situations. So beyond the ciminophil, we also looked into carbon tetrachloride and concannabinoid A-induced hepatitis. And as you can see here, the eosinophil accumulation occurred in both cases at 24 and 48 hours after challenge. And if we compare wild-type and knockout mice, we can see after carbon tetrachloride treatment, the eosinophil-deficient mice develop more severe liver injury. And also, if you inject ConA to the wild-type knockout, you can see that ALT levels are higher, necrotic areas are more extensive. So this seems to be a general phenomenon that eosinophils are recruited into the… when there's acute liver injury, eosinophils are recruited into the liver, and they play a protective role. So using the ciminophil model, we wanted to understand two questions. One is, how do these cells protect? And the other question is, how are they activated in the liver? In terms of how they protect, we had clues based on our experiments or our previous study in the ischemia reperfusion model. We found that these cells are the predominant source of IL-4 and IL-13 in the liver because they're known to be producing these cytokines. In the eosinophil model, we used IL-4 reporter mice, and we also did intracellular staining for IL-13, and we found that eosinophils are the main source of these cytokines. And my postdoc labs a long time ago had shown that if you knockout IL-4 or you knockout IL-13, those mice develop more severe ciminophil-induced liver injury. So we hypothesized since eosinophils are the main source of these cytokines, perhaps these cytokines are important in mediating the protective effect of eosinophils. So to test that, we generated eosinophil-specific IL-4, IL-13 double knockout mice. So in these mice, only the eosinophils cannot make IL-4 or IL-13. And so when we compare with the wild-type littermates, you can see the knockout mice have increased elevation of ALT and larger areas of necrosis, suggesting perhaps the eosinophil-derived IL-4, IL-13 play the protective role. Using it from another angle to test this, what we did was we used the Gadawan knockout mice. They have a very severe liver injury after a ciminophil challenge. So we give them either PBS or wild-type bone marrow eosinophils, which we knew what to protect the mice. You can see here in the middle bar that ALT dropped dramatically. But if we give them bone marrow eosinophils that cannot make IL-4 or IL-13, the protection doesn't happen. Liver histology showed a similar observation that the wild-type eosinophils can protect the eosinophil-deficient mice, but not IL-4, IL-13-deleted eosinophils. So these data suggest that IL-4, IL-13 production by eosinophils are important in their protection, their protective mechanism. Then we wanted to know how are these cells activated to produce IL-4 and IL-13. Again, we reasoned that there has to be something released during injury that triggers the activation of eosinophils. Again, borrowed from our previous finding in the ischemia reperfusion model, we know that eosinophils are probably the only cells that express the IL-13 receptor, and IL-13 is released as alarming during injury. So we wanted to first see if after semen phage challenge, IL-33 is also released. As you can see here, the blue bar shows a semen phage challenge. After semen phage challenge, we measured serum level of IL-33. It's dramatically increased comparing with PBS control. So we then wanted to see if IL-33 can stimulate eosinophils to produce IL-4, IL-13. And we know after semen phage challenge, there are other cytokines that's produced. Most typical ones are infrangamma and TNF-alpha. And we also use the combination of infrangamma and TNF-alpha to stimulate eosinophils. You can see that these are message levels of IL-4, IL-13. IL-33 can dramatically induce the message level of IL-4, IL-13. TNF-alpha, infrangamma combination, especially for IL-4, can also induce to a certain degree, but not as much as IL-33. Similarly, in terms of a protein, IL-4, IL-13 release into the supernatant, infrangamma, TNF-alpha can stimulate a little bit, but not comparable at all to IL-33. And we wanted to dig deeper to understand how does IL-33 ST2 signaling regulate IL-4, IL-13 production that we did not understand before. So what we did was, since we didn't have any clue, we used a screening approach called reverse phase proteomic analysis, RPPA. So we treated bone marrow eosinophils with just regular medium versus with IL-33, and then we sent the samples for RPPA analysis. And interestingly, the top candidate was cyclooxanase 2. And we confirmed that screening by qPCR and Western blot. You can see that if you stimulate the eosinophils with IL-33, COX-2 message and protein levels increased dramatically. And then we wondered, so we wondered if cyclooxanase 2 or cyclooxanases are activation is necessary for IL-4, IL-13 production. To do that, we blocked the COX-1 and COX-2 by endomethasin treatment. As you can see here, the open bars are eosinophils stimulated with IL-33 alone plus vehicle. But if we co-culture with endomethasin, we can see a dramatic suppression of IL-4, IL-13 release. To further confirm this, this is the case in vivo as well. What we did was we treated GATA1 eosinophil deficient mice with bone marrow derived eosinophils immediately before a simenophil challenge. And we have two groups of mice. One, we just give them wild-type bone marrow eosinophils. And the other one, we give them the wild-type eosinophils after incubating with cyclooxanase inhibitors. So you can see here, the ALT levels is significantly higher if we block cyclooxanase in eosinophils before we do the adoptive transfer, suggesting that cyclooxanase, perhaps through inducing the IL-4, IL-13 production, plays a role in the protective function of eosinophils. So I didn't get into further details, but our experiments and the data support this following hypothesis. What we showed was after a simenophil injury, IL-33 is released, which activates eosinophils. It first triggers P38 map kinase activation, which then is necessary for cyclooxanase-2 induction, which then triggers NF-kB activation. And in our hands, that becomes necessary for IL-4, IL-13 release. And the way IL-4, IL-13 protects was because it suppresses interferon gamma level in the liver. So with that, our second question was, how were these cells recruited into the liver? Our first hypothesis was that perhaps platelets were involved, because platelets are known to arrive very quickly whenever there's injury, and they are very important in recruiting other cells. Recruiting other cells, especially neutrophils. So we thought perhaps they're also involved in recruiting eosinophils. And so what we did was we depleted the platelets using anti-CD41 antibody, but we did not see any difference in terms of eosinophil number by flow or by immunohistochemical staining. So now our second thought was, in a lot of injury situations, neutrophils come in very early. And based on our kinetic study, eosinophils come in much later than neutrophils. So we thought perhaps neutrophils release chemokines or something, lipids, to recruit eosinophils. So what we did was we used anti-Ly6G antibody to deplete neutrophils. And again, we did not see significant changes in the number of eosinophils in the liver by flow or by IHC staining. So then we thought perhaps, we know that demonocytes were not recruited yet before eosinophils come in. So we thought maybe the resident macrophages are important. So what we did was we depleted the Kufr cells using liposome-entrapped chloronate, the CLD group here. The control were empty liposomes, the lipo group. As you can see, based on the percentage of eosinophil, percentage of Ccr3 positive eosinophils, there is a dramatic reduction of eosinophils by flow and a very obvious reduction in number by immunochemistry staining. So the next question was, how are these Kufr cells activated to recruit eosinophils? We reasoned that perhaps damage-associated molecular pattern molecules are important in this. We know there are several DAMPs that have been reported, IL-1-alpha, HMGB1, R2, most commonly most commonly reported. So we injected these factors to naive mice, just wanted to see if they can recruit eosinophils. But here by flow analysis, we did not see any change in eosinophil number. And then going back to our IL-33 as a larmin, we injected IL-33 to naive mice, and we can see significant increase in eosinophil number. And this seems to be macrophage-dependent because if we deplete Kufr cells before we give IL-33 injection, we don't see the increase. So if IL-33 is important, we would expect in IL-33 non-comized, if you give a simenosine, the eosinophil number should be reduced. And that is the case. Comparing with the wild type, you can see a dramatic reduction of eosinophils in the liver. And this is even though his chemical staining for MBP. So with that, we had two questions. One is, of course, how does IL-33 stimulate macrophages to recruit eosinophils? And the other question that we always had in our mind, even for our previous ischemic reperfusion study, that is, which cells release IL-33? So to answer this question first, we used the IL-33 reporter mice. And we found that surprisingly, it's not hepatocytes. It's the non-parenchymal cells. And so when we look at the non-parenchymal cells, we found about two-thirds of IL-33 positive cells are liver sinusoidal endothelial cells. And one-third are the Kupfer cells. But which one is important in IL-33 release? After a simenophil challenge, we're not sure. The only experiment we could do was we depleted Kupfer cells. And we did not see any difference in IL-33 release. That kind of indirectly suggests it's the endothelial cells. But we don't have direct evidence. So then to get to the other question, how does IL-33 activate macrophages to recruit eosinophils? We first looked at eosinophil chemoattractants. There are two major ones in mice. One is eotaxin-1, CCL-11. And the other one is eotaxin-2, CCL-24. And you can see here when we measured the message and protein levels in the liver after simenophil treatment, we did not see any difference in elevation of CCL-11. But there is a significant increase of CCL-24, both message and protein level. And if we deplete Kupfer cells, the CCL-24 level is reduced. So it looks like the macrophages through producing eotaxin-2 to recruit eosinophils. Now the question is how? The first thing we did was we asked if IL-33 can directly activate macrophages to produce CCL-24. And the answer is no. If we take bone marrow-derived macrophages or Kupfer cells, we treat them with IL-33, there's no increase of CCL-24. This kind of is consistent with our operation throughout is that in the liver, the only cell we can detect IL-33 receptor on are the eosinophils. So we thought perhaps the IL-33 still firstly activate eosinophils and then release something to activate macrophages. So to test that, we took the liver mononuclear cells and we use MaxBs to purify or enrich the eosinophils. In this portion, we get 85% of the cells are eosinophils. And we also kept the non-eosinophil portion. And we co-cultured bone marrow-derived macrophages with the eosinophil enriched portion and eosinophil deleted portion. As you can see here, if we culture the macrophages with eosinophils in the presence of IL-33, there's a dramatic increase of CCL-24 production in the supernatant. So this suggests that the eosinophils triggered by IL-33 is necessary for macrophages to produce CCL-24. And we know that we did an experiment with a transwell culture. And we know that this is not cell-cell contact dependent. It's a soluble factor. So here we showed that if you take bone marrow-derived eosinophils, treat with nothing or IL-33, and then you take the supernatant from these cultures to culture bone marrow-derived macrophages, you can see that the supernatant from IL-33-stimulated eosinophils can trigger macrophages to produce CCL-24. And then we wondered what's in the supernatant. We used several different antibodies. And surprisingly, we found only IL-4, but not IL-13, mediate this function. So if you block IL-4, the macrophages in the supernatant, macrophages cannot produce IL-24. But anti-IL-13 antibody doesn't seem to do much. And another interesting observation was that if you culture the macrophages with this IL-33-stimulated eosinophils, after a few hours, you cannot detect the IL-4 level in the supernatant decreased dramatically, suggesting it used up by macrophages. So I want to summarize this part of the talk. We found that after acute liver injury, especially overdose of a ciminophil challenge, IL-33 is released, perhaps from liver sinusoidal endothelial cells, and that triggers the activation of resident eosinophils. And these cells then produce IL-4 to activate KUFR cells to produce CCL-24, which then recruit a lot more eosinophils. We do see, by flow cytometry, there is about 2% to 3% of CD45-positive cells are eosinophils. But after the injury, a dramatic increase in number of these cells in the liver. So with that, I'd like to thank my team here at UTHealth. This work I discussed today was done mainly by Yang Yang and Long Xu, who's left the lab, and also my collaborators at UTHealth, Mayo Clinic, University of Colorado, and NIAAA for their generous support in providing patient samples, mice, and bouncing off ideas, and also funding from NIDDK, as well as the STAR award from Texas. Thank you very much for your attention. Wonderful talk. Can you hear me? Yeah. I had to switch this. Fantastic talk. Thank you. Thank you so much, Cynthia. There's a few questions, I think. Adayan, you go first. Hey, Cynthia. Can you hear me? Okay, there you go. Oh, hi. Great talk. This is really amazing work. I have a question about IL-33 or related cytokines. In the initial part of the talk, you showed patient samples, and you showed that the ones with acetaminophen injury have eosinophils, right? And I'm assuming that most of those are either patients that need a transplant or that died, because that's how you get the liver. Those who survive, you don't get a biopsy, right? Yeah. And so that's a big problem in the field, anyway. So I was wondering if there is a difference between eosinophils in patients that survive versus they don't. And I know you can't look in the liver. I don't expect you to know that. But can you measure something in survivors versus non-survivors that's a pseudomarker of eosinophils in the liver, and that can inform you about it? Yeah, that's an excellent question. Thank you, Adayan. We struggled with this question, but just like you said, I tried very hard to get some sample. There's not much we can measure in the blood. It was impossible to get liver biopsies. I can tell you, in terms of liver transplantation, we could get, not for very long, but right before they close the patient, we get the sample. So that's about two to three hours after the surgery. So we measured eosinophils in those liver transplant recipients. And in our previous paper, we found that the number of eosinophils correlate with how well they do day one after surgery in terms of the drop of ALT. So in the same context, did you or anybody else has gone to look at, say, the radiometric volumetric of liver transplant and see if that correlates with eosinophils or something like that? So you could look at how the graft grows later, regenerates later in the transplantation, and you could correlate that with some of the eosinophil markers in the blood. Has anybody looked at that? Yeah, but that's again something we cannot, you mean, we've been trying to get liver biopsies several days after transplantation, but that's impossible. I wouldn't, yeah, no, no, I wouldn't get the biopsies. They can do radiological analysis. So this is non-invasive and, you know, you can get a sample from the same patients. Of course, I think there is a, just like acute liver failure study group, there is another study group that studies that, and they might have samples like that. So we'll look into that. And also with acute liver failure study group, I mean, Lily knows this too, there is a lot of samples that you can get from survivors versus non-survivors. And maybe you can look into that, see whether any of the cytokines that trigger eosinophils, they kind of, you know, come up in a particular population, I would say. Yeah. So unfortunately, I think we have to be more creative. IL-33 is not a good cytokine to measure. It's not very robust in terms of, yeah, we have to maybe, and I don't know how, you know, usually the eosinophils don't dwell in the circulation. So it's, yeah. Thank you. Thank you. Wenjing, you want to ask your question? Oh, yeah. Hi. Hi, great talk. Thank you. So you have a question about the mouse experiment. You have the ALT data, should eosinophil, actually could very early, when you manipulate eosinophil numbers, it protect early at 8 hours ALT and the later 10.24. I think the early phase probably still undergoes narcosis. My question is, how do this eosinophil affect with well-known, you know, protein adducts, mitochondrial damage, narcosis, maybe even junk activation? How do this eosinophil get, you know, impact on this, we know, cell death pathways? Yeah, that's a good question. We didn't look into that. But my prediction is the eosinophils don't have any effect on those. I don't think the protection of eosinophils is attenuating the direct toxicity of ciminophil. As you can see, when we do the anti-IL-5 treatment, and also when we compare wild-type and knockout mice, we did not see difference in 8-hour ALT. I think the 8-hour ALT is reflection of the direct toxicity of NAPQI. But so, but I think the eosinophil, the eosinophil act on the exacerbation. We're still trying to figure out what is the, what is the, you know, what I mean by, I don't know what I mean by exacerbation yet. We need to figure out, is it spreading of hypothesis injury, you know, maybe, maybe two layers of hypothesis are injured by NAPQI, but then the other, the layers outside is caused by the exacerbation. I think that's what maybe. It makes sense. Look at your staining, mainly kind of surrounding the necrosis area. Exactly. It's always in the necrotic area. Thank you. That brings us to Ben's question. Go ahead, Ben. Yeah, Cynthia. Well, I just lost the connection. I saw your one slides, you know, in your APP model, all these neutrophils accumulate inside of a necrotic tissue, lesions. But you also have a Connelly model, seems, you know, it's not that many, but surrounding the necrotic lesion. In the Connelly model, there's a necrotic tissue inside, it's not that many cells. But in the APP model, you know, we also stain, there's a lot of Kupfer cells, stellar cells, even endothelial cells. You know, I have a question, you know, whether, maybe you can find the slide, you know, what's very interesting. In the APP model, you know, it's a very high dose. Can they also kill all of those cells inside of a necrotic tissue? Sorry, what? I see many of them, if you do high dose. The mice will die. Will also destroy, yeah, this one. Yeah. Yeah. If we see the top panel, most of this is inside the necrotic tissue. Yeah. But the Connelly, it's, yeah, on the border. Yeah. You know, I have a question for, you know, the, you know, well, you did a carbon tetrachloride product similar to APP. Yeah. Oh, sorry, this is a carbon tetrachloride, you know. This is, but I can find APP. But we can see here, it's, for Connelly, it's definitely on the border. Yeah. And this is a, this is a patient, it's different. So this APP model, carbon tetrachloride model, the necrotic lesion still have structure of, you know, sinusoid in the flow of blood. I, I don't know. How do you, how do you measure? How do I measure? I'm just, I'm asking you. No, the Connelly model, clearly, everything inside was destroyed. I mean, there's no endothelial cells. That's, you know, that's made some few died. There's no corpus cells, but in Connelly, in the APP model or carbon tetrachloride model, if you're standing, you do see endothelial cells. We do see the corpus cells. But when I ask, what's the name? Frank Taki. He said, well, if you use very high dose, they may also destroy everything inside. Is that true or? Well, you guys all do APP models. I don't know. Ludayan, you use high dose. Ludayan. Yes, ma'am. You, you, you use high dose to, to use, if you use very high dose of acetaminophen, do you still see intact blood vessel? Yes. Yes. So what happens is with 600 dose, what we have generally used with about 12 hours fasting, you will just, you will get about 40% necrosis, 40, 45% necrosis. So it's pretty bad, but zone three is almost always intact. So, so not zone three. So zone one is almost always intact. So peripherally, you will see, you know, you'll see standard architecture. You don't see a terrible difference. And the necrosis is quite like, you know, just like an acetaminophen, it's like a zone. But in, in the necrotic area. Oh, and it's all blown up. There is nothing there. In the necrotic. Acetaminophen is a toxin to L-sex. I think Laurie's showing us. Yeah, it is. It is. Yeah. It will kill, but it will kill within the necrotic zone. It won't kill those away from the necrotic zone. It won't kill those within the necrotic zone. Within the necrotic zone, it won't kill those away from it. So in the necrotic area, everything will be destroyed. I don't think you'll see much there. All cells will die, hepatocytes and non-parenchymal cells. But Ben was saying that he can still, in the necrotic area at lower dose, like 300 mg, he can still stain endothelial cells and Kufr cells. Yeah. In the lower dose, you may get specific hepatic necrosis that is not spread out, but at higher dose, it is so such more injury that it kind of spills out. And I think it, it will kill other cells too. So the mice can survive? In our hands, we get about 20 to 30% mortality with 600 dose. Depends on the substrain. Neil is also an expert on this. Depends on, you know, N, J, how well they're- This is all about C57J. This is all about J and male mice. If you have C57J female- N definitely will die with 600 dose. Yeah. N are very sensitive. They will die at 300 dose actually. And so this is all talking about C57J and male mice. Females are resistant, but female mice are not very different at 600 dose than male mice. At 300 dose, they are different. 600, they are not different. And Hartmut just published that last year. So it is a function of dose, of course. I think one of the reasons we use the high dose, Cynthia, is to kind of mimic the survivors, non-survivors. Well, you also destroyed the central vein. Yeah. We sometimes see the debris kind of fall in and you barely see visible central vein. It's all weird. Mouth stuff. I'm more like the most thorough- I guess it's about the napkin, right? Once the napkin spills out, it can be toxic to anything. Exactly. I don't see any other questions, but I have one and it's, I don't know how to phrase it because, so when we think about eosinophils, normally, we think about type one hypersensitivity reactions and allergies, right? And at least in the Dili world, when we think about eosinophils, just what comes to mind is dress, right? Drug-induced eosinophilia, right? Rash. And the mechanism of injury in idiosyncrotic drug-induced liver injury, right? If you think like phenytoin or, you know, a drug that causes dress is adaptive immune-mediated, right? So it's a completely different model. It's hard to model it, obviously, in animals than these direct toxins. And I know you're an expert in like, you specifically focus a lot on innate immune cells and not necessarily on adaptive immune cells, but do you think that the eosinophils in those idiosyncratic Dili's are also there to protect, or do you think it's something completely different that's going on when you're talking about adaptive immunity and idiosyncratic Dili? Yeah, I think that's a very interesting question. I don't know. I mean, the dress seems to me it's an allergic reaction. So in terms of adaptive immune response, it may be a Th2. If it's a Th2, eosinophils are definitely critical in terms of its release of IL4L13 initially. It triggers the Th2 T cell response, which makes a lot more IL4L13. So I think the eosinophils do interact with T cells. On one hand, if it's a Th2 response, eosinophils will enhance Th2 response. And the other thing is we found in vitro, at least, that eosinophils are actually important in promoting cytotoxic CD8 T cell response in terms of the production of interferon gamma by CD8 positive T cells. The mechanism, we don't know yet, but I think the models we are doing now, they don't involve adaptive immune response. So in those cases, eosinophils, it has been shown in other tissues as well, that eosinophils has this function of tissue repair and remodeling. I think maybe that's why whenever there's acute injury, they come in. But when there is adaptive immune response, it depends on what type of T cell they interact with. I think you can worsen or enhance Th2 or promoting Th1 even. Like in cancer situations, people have shown, it's controversial, but in certain cases, eosinophils correlates with better prognosis, perhaps because they promote anti-tumor T cell response, like CD8 positive T cell response. It's very interesting. It's worth studying because there are, skip, go ahead. There are samples of that. This is very interesting. Eosinophilia is well described with immune checkpoint inhibitor toxicities and things, including, I believe, liver. We've seen that. And my question to you is, does it make sense of the checkpoint inhibitor? Are they just releasing eosinophils as well? And they're just a part of that? And is that why maybe the liver injury with checkpoint inhibitors is, despite the enzymes being sky high, the prognosis is pretty good. I know we step in with steroids, but in general, they don't get jaundiced. It's not that bad. So I just wanted to know that my specific question is, is the eosinophil that they're seeing with checkpoint inhibitors, is that just part of the general immune release that these checkpoint inhibitors, does that make sense to you? Or do you think it's in response to these injuries where the eosinophils are coming in and trying to repair things and protect things? Okay. Oh, that's very interesting. I didn't know that with checkpoint inhibitor injury, you have eosinophils? Yeah. I think there are all these reports with pneumonitis and all these other things. And I think even when you don't have any injury, I think there's some data where they have the eosinophilia in general, but it's all evolving literature. But yes, there are lots of publications about that. Yeah. So your question is, are they protective? No. Do you think it's makes sense to you that the checkpoint inhibitors are just allowing eosinophils to come up along with other, it's just a general response to removing the checkpoint. Is that what's going on? Or do you think it's the eosinophils- Because they express CTLA-4. Yeah. Or is it just the eosinophils are actually adapting and they're coming in after some reaction and they're trying to protect things. Do you know? So I think, do you see a necrosis, necrotic injury? In the liver. Yeah, sure you can. Yeah. So I think if it's necrotic injury, I would predict it's because of the injury they come in. If there's no injury and you see this, usually it's because of started with ILC2 cell activation, which is not a lot in the liver, but maybe if there's a, for some reason, there's a TH2 T cell response, then you will see eosinophils like allergic lung or surgeon lung situation, which is very TH2 oriented. Your eosinophils can be amplified. But I think for the liver, unless there's a situation that causes, maybe checkpoint inhibitor treatment causes TH2 response. It does. It does? Yes. It activates TH2 cells or- It activates a lot of things including TH2 cells. Yeah. Yeah. And it's interesting because you also mentioned that they express SCATA and that's also one of the checkpoints. So I wonder if that also gets like kind of upregulated or simulated when you're blocking the other checkpoints. I mean, there's a lot of, it's complicated, right? There's a lot of interaction between innate and adaptive and between all these cells, but yeah, TH2 cells do go up. And there's actually, at least in my model, I see IL5, I see IL4. Okay. So in that case, this is a stew of TH2 cells and mediators. Yeah. Then I would say this is not, I see eosinophils as this, I put them into two bins. One is they are there to repair acute injury. And the other is they're there as a part of the TH2 response. But surprisingly, even in asthma, people have shown with the knockout mice that they are not key in causing asthma. You get rid of eosinophils, the eosinophil knockout mice, the asthma severity does not change. I think they are, that once you have TH2 cells, you don't need eosinophils. But they are important in initiating or part of the initiator of the TH2 response. Thank you. Interesting. Yes. So this is such a creative approach. It's very interesting. Thank you. I think I speak for everyone that it's fascinating because it's not something you normally think of eosinophils and the data is very compelling. So I don't know if there are any other questions. Great. We're going to be right on time. It's nine o'clock. Thank you everybody for joining us. Thank you, Dr. Ju for sharing your work. Thank you, Dr. Ju. That's great. Thanks. Thank you. Bye-bye everyone. Have a good rest of your day.

eosinophils

liver injury

acetaminophen

immune modulation

cytokines

IL-33

liver repair

therapeutic approaches

drug-induced liver injuries