International Medical Conference Endometriosis 2025:
Endometriosis 2025: Your Mother Should Know, Your Doctor Should Know Better!
Transcriptomic Changes in Superficial Peritoneal Endometriosis - Asgi Fazleabas, PhD
My next speaker is going to be Dr. Osky Albas. This is going to be on Transcriptomic Transcriptomics. He's going to tell you what it's going to be about. Dr. Albas is the distinguished professor and associate chair for research in the Department of Obstetrics and Gynecology at the University of Michigan. His groundbreaking research focuses on uterine biology implantation and the pathobiology of endometriosis.
Good morning everybody. I want to thank Tamara and Dan again for having me here, and I always enjoy coming to this conference and some of the work that I'm going to tell you about is really recent. We just published it and really it had a beginning from the funding that we got from the EFA to get this started. So basically, oops, I'm missing a slide again. There we go. Yeah, keep skipping, right? I don't have to introduce this slide to this audience of course, but I just want to set the stage. So what we've been trying to do is, as all of you well know, superficial peritoneal disease is one of the most common types of pathology that we see in so many of our patients. And yet, because of the difficulty of getting enough tissue and begin to try to understand what might be happening in the development of these, we've been trying to focus our work on looking at the molecular changes that we see in the superficial peritoneal disease.
So these are studies that we've been doing recently in trying to understand what might be happening at the transcriptomic level, particularly with regards to the superficial peritoneal lesions. All right, we'll keep having to go back. Okay. So this is a initial analysis we did. This is with bulk RNA-Seq analysis. And in terms of looking at the inflammatory characteristics that we find when we analyze the peritoneal lesions, comparing with the utopic endometrium, you can see on the walk volcano plot on the right hand side that there's a large increase in the number of inflammatory genes that show up in these peritoneal lesions when you compare them with the eutopic endometrium and with gene enrichment analysis of the hallmark pathways. Again, you can see that many of the inflammatory pathways, particularly the TNF alpha pathway and a number of the inflammatory responses, appears to show up with regards to the changes when you look at this whole compartment and the whole complex within these lesions.
Well, the issue here, of course, is that when you're looking at the lesions and when you're doing bulk transcriptomic analysis, this is a slide that my postdoc who's really done all the work Greg Burns has done. So when you think about a bowl of fruit and you think about doing bulk sequencing, what it is is you're looking at, essentially everything is all mixed up and you're looking at a slushie and you really don't know outside of the color and what it contains. You don't know really what's in the slushie, so you have to make analysis to try to identify what they are. Single cell RNA-Seq is really good, but again, you essentially take these single cells, you analyze them, and it's almost like you're tasting each one separately blindfold to try to see what's in there. The advantage of spatial, of course, is that you have a context of each single component and you're able to then analyze each single component and see what's within this component to be able to try to get ideas about what might be happening in terms of development of these lesions.
So in response to trying to do that, we wanted to find out first, what are the alterations that we see in the peritoneal lesions? Are there specific cell types? And what is the crosstalk between these different cell types that we see within the lesion? To be able to try to understand if there are communication pathways that we might be able to try to begin to understand the inflammatory components that's found within this lesion. Oops, sorry. Keep going back. Maybe I push this slowly. Okay. So this is the group of patients we did. They're not ideal. As I said, this is preliminary data. As you can see, the older patients, they have other pathologies. But the advantage we had with doing these analysis was we got samples from women that were well characterized. They were later in their, they were older, they were going through hysterectomies and in addition end endometriosis.
They also had other pathologies, but the advantage was that they were taken all at the mid secretory phase and we had matched endometrium and peritoneal lesions from these patients and we then subjected them to spatial transcriptomics. So essentially for this study, we use the NanoString platform of geos, and what we can do is take these lesions and you can see this is the endometrium and the lesion. They have both, and we selected within the lesion regions that had both epithelium, stroma and macrophages. And by staining them, by using the different markers, you can then isolate each of these different cell types. And what we asked was to ask, identifying these different cell types and using the NanoString platform by isolating these cells, we wanted determine what the differential expression was by cell type, as I said, and try to see whether there is an altered cell to cell communication.
And when we did the analysis, when we first looked at the epithelium, what again we found was that there was again, a large change in the number of immune parameters within the epithelium itself, which then again with gene hundred print analysis, you can see again by the size of the dots and the analysis that there is very positive adaptive immunoregulation going on within these epithelial cells. And one of the ones we see, sorry. Okay, let's go back again. I skipped that slide. Okay. Alright. Do you want to change them? That'd be easy. Okay. So with the spatial transcriptomics, this one was an important one that I missed. If you then took the dot plot analysis and confirmed each of the different transcriptomic changes within each cell type and looked at the different genes that would identify each of these different cell types is the epithelium, the stroma, and the macrophages.
You can see that they were clearly separated out with spatial transcriptomics based on marker genes that you find that identify each of these different cell types. And the other advantage is that when we look at the umap analysis of these samples, you can see that within each patient they clustered together, which told us that the endometrium and the were coming. So the lesion was reflecting what's happening in the endometrium and that each patient was its own sort of self-control. So we are able to then compare the utopic endometrium from the ectopic lesion. So the next slide, and then when we go ahead and did all of the analysis using our analysis and this paper that was just published, those of you that are interested in the painstaking bioinformatic analysis that go on, it's all described in the supplemental in this paper. Again, what we did was we were able to then see that even the epithelial cells themselves have a large complex of immune responses, which can then be further analyzed using gene ontology analysis.
And that there was a very positive response in terms of the immune changes that we find in the epithelium. Next slide. And one of the ones that we were highlighting is C3. And as I go through the talk, you'll see why we've kind of picked out C3 initially to try to do these studies. Next one. And then again, if you looked at the macrophages themselves, which again that we were able to separate out, we saw that there was an increased inflammatory response in the gene sets that really identified the recent differentiation of these monocytes compared to the tissue resident macrophages that are found in the endometrium and that the receptor for C3 was markedly increased in these monocytes and these macrophages. Next slide. And again, the gene endocrin analysis tells us that we see the same sort of changes that I showed you with the bulk RNA C analysis that I showed you on the first slide.
And we see very, very similar pathways that show up in these macrophages that we've isolated by spatial analysis. Next slide. And then if you go back to make sure that these are of peritoneal origin compared to eutopic endometrium, you can see here again the tissue to resident macrophages are largely expressed in the lesion and also the activated macrophages are largely expressed in the lesion compared to the utopic endometrium. Next slide. And if you look at the peritoneal markers for these macrophages, here again, you find that the peritoneal macrophage markers are only found within the lesion and not found within the eutopic endometrium. And those, a few that read this literature, you know that the macrophages within the lesions are very, very different in terms of their ability to maintain the lesion and not do their normal job of moving out or clearing the debris within the lesion.
So essentially what you're seeing is a change in the type of macrophage function within these lesions that might be contributing to the persistent development of these lesions and not removing the lesions or removing the debris that comes back with retrograde menstruation. Next slide, please. So we've seen these changes within the epithelium and we've seen these changes within the macrophages. And how do they then compare if we try to understand what the crosstalk is between these different cell types? So using ligand receptor protein analysis, what we can see is that the stromal cell surprisingly had very little interaction with any of the other cell types. In contrast, what we see is that the epithelium and the macrophages really have the most persistent interaction with one another when we take the analysis and subject them to ligand receptor analysis. And here again, you can see that the complement pathway that I identified is highlighted in this interaction between the epithelium and the macrophages.
And you also see good interaction between the macrophages themselves using cell chat. We then try to understand what the role of the complement system particularly was. And if you then use this using cell chat, you can see that the interaction between the epithelium and the macrophages in the eutopic endometrium was relatively minimal. But if you look at the interaction between the epithelium and the macrophages within lesion, you really see a very strong interaction between the complement system and between the epithelium and the macrophages, particularly related to complement system. Next slide. So what we've been able to show, at least by transcriptor analysis, we can see that there is an inflammatory pathway that seems to be associated with the complement pathway and that this complement interaction from the epithelium and the pro repair macrophages might be important in promoting this pro repair macrophage phenotype that are pro lesion.
And that might increase as these lesions develop. So to be able to try to understand what the role of this interaction might be, we are now moving into doing some other experiments to try to understand this interaction and see whether they really do play a role in terms of angiogenesis, fibrosis and a number of the other pathologies that we see with these lesions. So next slide. So what I've tried to show you right now is that we see an increase in immune response genes in the epithelium. There's a shift in the subpopulation of the macrophages and that the lesion epi epithelium does contribute to some of the inflammatory signaling pathways. Next slide. So a hypothesis right now is to ask whether the endometrioid epithelium promotes a local inflammatory environment that is important to sustain the peritoneal lesions. So our analysis that we are now doing in collaboration with Stacey's beautiful collection of her A two A adolescent patients, and comparing them with our baboon models.
So those of you that have followed our work know that for many, many years we've developed the baboon as a model to try to understand peritoneal disease. And what we're trying to do is use the analysis and the samples that Stacy has along with Katie Terry to look to see how we can compare our baboon lesions obtained at about 12 to 15 months after the induction of the disease because we know exactly when the disease was induced and compare them with A two A cohort where these young women were diagnosed with disease 12 months prior to surgery. So next slide.
So when we look to see what happens with regards to the lesions and compare the lesions in the baboon and humans, we see that they have very similar pathways within the subset of the analysis when we are comparing the baboon and the human. And we see a number of different hallmark gene sets that are common to both, as you would expect with EMT inflammatory response and KRAS signaling. And that we can very clearly differentiate using principle component analysis, the genes that are found within the endometrium compared to those that are found within the peritoneum. So there, there's a nice separation between these two. And then if we then go to the next slide and superimpose the analysis that we get from the young women's lesions with our baboon, you see that there's very significant overlap that tells us that there is a number of different common genes that we can see within the lesions from these young adolescent girls with our baboon model.
Next slide. And then when we go back and look at our cell chat, and more recently, Stacy's group has done single cell analysis from these lesions from the A two A group. And you can see that the myeloid cells separate out beautifully right here with single cell analysis. And this is just comparing it to a search at analysis. And then if you look to see what genes are selectively expressed, you can see here, again, complement three and its receptor are again separated out very nicely between the lesion and the epithelium, just like in the adolescent lesions compared to what we had seen in our single cell in our transcriptomic analysis. And then if you look at the interaction between the receptor for complement three and CD 68, which was the marker that we used in our spatial transcriptomics to identify the macrophages to do the analysis, you can see that they're co localized very nicely in this single cell analysis when you look at the analysis from the adolescent lesions.
So now the question is next slide. How do we begin to try to understand what the crosstalk is, what the interaction is between these two? So a couple of years ago, we developed a steroid system where we can create what we think might be an in vivo component where we combine epithelial stromal cells and create these endometrioid that migrate through a maal matrix that contains the LP nine cells that are found within there. And then you can see that over time this complex, and particularly the stromal cells that are found in the center part of these sphe, migrate into the matrix containing the peritoneal cells. Next slide. And again, when we compare the analysis of these OIDs together with the baboon and human lesions, there is again, a lot of interaction between these three showing that the sphe are reflective at least of what might be found within these peritoneal lesions.
Next slide. Then we ask the question, if you do a very, very simple experiment and we co incubate the sphe with macrophages, and these are just a large number of macrophages in the media, you can see that in the presence of 7,500 or 15,000 macrophages that there is an enhanced migration of these spheres into the magel layer compared to the controls. So the presence of these macrophages in the media seems to enhance the movement of these lesions into these s at least into the matrix and enhances its invasion process. So these are the kind of tools that we want to now begin to use and try to understand how the crosstalk and that we find between the epithelium and the monocytes and the macrophages might be playing a role in contributing to early lesion development. So the last slide is really another component that we are beginning to look at.
And we are working with our colleague, Ben Chen at Michigan State University, where he's developed this logarithm where he has an entire database of all of the FDA approved drugs that you can then plug in any of your pathways that you find to find if the pathways you've identified in the context of endometriosis, for example, could identify potential drugs that could be used to try to analyze and use them for non-hormonal therapeutic analysis. So this is the pipeline, oops. This is sort of the pipeline that we can use. It's a very complex pipeline, but you can find a number of different hits and go down the list and narrow it down using our different model systems to see whether we can get initial hits of non-hormonal therapeutics that we can potentially use, and then transfer them back into our in vivo model systems to see whether they have any use or any help in preventing at least lesion development or lesion formation. So with that, the last slide next, so really thank the people that have done the work. I mean, this was all led by Greg Burns, who's an amazingly brilliant mathematician. I mean, half the time, I don't understand what the hell he's talking about.
And then the PHE system was developed by Yang. It was another postdoc in the lab, and we wouldn't have been able to do any of this work without our collaborators, Italy Flores, who helped us with the initial studies giving us the samples. And our continuing work with Stacy and Katie. Erin Greaves has been an enormous contributor to our understanding of macrophages and my colleague Aurora, and we're thankful to EFA NanoString and of course NIH. And Stacy and I now have a grant sitting that should be funded in under normal circumstances, but it's sitting there in IH right now to continue all of this work. So thank you very much.
