International Medical Conference Endometriosis 2025:
Endometriosis 2025: Your Mother Should Know, Your Doctor Should Know Better!
Transcriptomic Mapping of Endometrium and Endometriosis - Semir Beyaz, PhD
Last but not least, we have Dr. Samir Beyaz. He is an assistant professor at Cold Spring Harbor Laboratory in New York where he heads a research group investigating the cellular and molecular mechanisms linking nutrition to health and disease. He earned his PhD in immunology from Harvard in 2017 focusing on the dietary regulation of intestinal stem cell function and tumor genesis. Dr. BA's BA's current work explores how dietary and metabolic perturbations affect the immune system and contribute to disease associated with immune dysfunction such as cancer.
Alright, thank you so much Dr. Chu and thanks to Endo found for the invitation. It's always great to be here. As Dr. Chu mentioned, I'm coming from a different field. I studied immunology, blood cell development part of my PhD and then I moved to the gut to study epithelial cells. In that context, I learned a lot about your body doesn't care whether you did your PhD, your is just work in harmony and it's up to us to figure out the language of life. And this will be the main discussion that I want to have with you. And it's great to follow Linda and because they did a great introduction to many things that I will highlight. So in my lab we focus on understanding how nutrients and metabolism, both at the organismal level and at the cellular level, influence who we are. And by saying who we are, it means that are we healthy, are we diseased?
And what are the mechanistic determinants of that process so that we can have better diagnostic, preventative, and therapeutic tools. And so for endometrium, which I serendipitously entered that field and I'm glad I did that, we need really better fundamental approaches to understand how this tissue and organ works and operates in human setting. And so echoing what Linda said, I did my PhD generating mouse models of diseases, genetically engineered mouse models. I still use them. They are valuable tools to understand causal mechanisms. We need those tools. But for tissues and diseases of endometrium, we really need to have a better understanding of human uterus. And by better understanding, I mean we really need to know what are the cells, how do they look like, what do they produce? We have historical understanding of that tissue, but the depth that biomedical community has in terms of the genetics and the pathways and the interaction between multilayered regulatory networks that involves chromatin metabolism and signaling pathway is so advanced yet in your tissues, like endometrium are always lagging behind.
And I can tell you, because I study both colon and colon diseases and endometrium and endometrial diseases, there's a gap around like 10 years. People started applying some of these methodologies and approaches almost more than 10 years ago. And I was part of some of those studies. And so I'll pause for a second and why is this? I studied colon, I study colon diseases and then there is endometrium and endometrial diseases. So there are many, I'm preaching to the choir here because you guys have been dealing with this disparity way more than I was thinking. But I think more than ever these days we need to speak this out loud. And as someone who entered this field very recently, I'm very grateful for how welcoming the endometrium and endometriosis and overall the gynecology community is. I was talking to Dr. Hugh Taylor earlier and he said, Colin is really, really good too.
Column field is not very welcoming. Endometrial field is much more welcoming than the column field. So thank you for that. So coming back, why as a mechanistic scientist, I'm highlighting to you, I'm preaching to you that we need better models because at the end of the day, if you look at mouse and human, they're not different. And I love Brian Kim's article as well that Dr. Griffith mentioned earlier. As scientists we have to be reductionists. I'm a molecular biologist, we have to build models. But sometimes scientists become so obsessed about the model that they have in hand. And because it is easy, they do everything with that model. So at the end, they cure a condition that does not have any relevance to the actual intention. So bottom line for me, and I think for majority of the community of studies, endometriosis, if you want to cure endometriosis, if you're not understanding endometriosis, you really need to go to the human.
So we have been developing some approaches to first characterize endometrial epithelium and other components of the complex cellular network that it has from humans because there is a gap. And this gap is not just profiling five women. Endometrium is a dynamic tissue. It changes drastically every month and across the lifespan of a woman. And then as Linda highlighted, we really need human models to study the biology both in health and also in diseases. And if we cannot do both of these, we will still be lagging and not making actionable progress. Good news is that many groups now are paying attention to that problem. My group, alongside with several other good collaborators in New York City area, came to together now almost five years ago, and we wanted to create a very comprehensive atlas of human endometrium across women from different demographic aspects and also stages, premenopausal, postmenopausal, different cycle stages.
So that was a very big undertaking at the time. I will tell you, when we applied for this project that is now funded by Chan Zuckerberg Initiative, we were one of the few groups that said, we want to do something about endometrium. Just to highlight to you again, this disparity now after a lot of ups and downs, a lot of learning, a lot of failure. I'm happy to tell you that now we have successfully generated a comprehensive atlas of human endometrium from 62 patients. So that was a very big undertaking. We recruit more than a hundred plus patients into that study, but it took us almost two years to fail miserably to get the assay work. And once the assay work, then magically 62 patient got profiled. But that's just the highlight. Failure is important, especially in areas of research in which we need more patients and we need more funding.
I know I'm again preaching to the choir, but we need more funding for women's health and endometriosis, especially this day and age because these type of assays are extremely, extremely expensive. I mean that. So the emphasis in that study is that there is also a disparity when you look at biomedical enterprise that we tend to profile people who are able to afford to go to hospitals that can do such experiments. And the profile of these people are usually white. So most biomedical research enterprise is studying diseases that affect white men. But some places in the world who care about women's health are also profiling white women. So when we went to the Chancellor Berg Initiative, it was apparent that if you look at all the data repositories of single cell atlases, almost 90% of these gene expression data cell state data is coming from white people.
Nothing wrong studying white people, but if you just study one genetic makeup of a diverse human population, you are really missing out, right? We are humans, we are, this society is amazing because of all of the colors that it pertains. So this is what we did. We took into account self-reported ancestry across different stages of women. And then we have done whole genome sequencing because people lie. People can say that, Hey, I'm Native American, I'm Hispanic. So we also did low pass whole genome sequences and then identified some of these ancestral SNPs that gave us this nice distribution. So this is how our current dataset looks like. We profiled human endometrium by looking at the gene expression, which poly explained that it's true single cell RA sequencing. And from the same cell we have done chromatin accessibility profiling using ATAC sequencing. And then we have this nice harmonized RNA and ATAC profile that can tell us now what these cells look like.
So this is what it looks like. Now we have different cell clusters that contain different flavors of epithelial cells, stromal cells, endothelial cells, immune cells. And we are now stratifying these based on different type of phases. And some of these cell types and names and states make sense like when you are going from one phase to another in a woman group that's premenopausal, the changes make sense. And then we also are now trying to compare these cell state and cell stages between premenopausal and postmenopausal because we have enough sample size to do comparative analysis to understand how does this tissue adapt towards these changes across the lifetime of the patient. And so one of the most exciting parts for me in that study is that it'll allow us to go from descriptive, I profile these genes, and this is how it looks like to ask mechanistic questions.
Why does a cell behave as it is in the endometrium or beyond? Mostly dictated by what transcription factors are activated. And thank you Palona for explaining transcription factors. Transcription factors bind to the DNA turn on genes and those genes are then become protein or become regulatory RNAs to carry out important cellular functions. And so here this is showing you the map of chromatin accessibility across all of these women that we profiled amongst the cell types that we identified. So why is this important? It is important because for us to understand health and how does a healthy endometrium looks like and when does a diseased endometrium become a disease, we really need to understand these master regulators. So this is a very exciting data for me because I spend a lot of time thinking about transcription factors and how they sense environment and how they define health and how they influence disease.
And there are lots of interesting transcription factors. I don't expect you to recognize any of them, but if you do, I would love to chat with you that are defining the activities of them, defining these cell types. So Palona mentioned gain of function, loss of function, experiments like you either delete something or activate something. And so the next stage for us, now that we have these patient samples, and I also mentioned to you that we develop the organoid models from the same set of patients, the next immediate experiment for us is to either knock out or over express to see that if these master reor transcription factors are actually sufficient to drive these epithelial cell states. And so because we have a comprehensive patient group, we can do experiments that you may be curious about. What's the difference between pre versus post in these cell types?
What persists? What dissipates? So this is another analysis that can tell you which transcription factor in comparison in comparison between post versus premenopausal endometrium are enriched in their activity within these cell types. And so this is important for people who have ambitions to enhance endometrial receptivity or its function beyond what is possible today. And transcriptional regulators are very important in endowing cells, these type of features. It's a larger effort, but this is where this type of foundational data sets can lead to. So last but not least, I highlighted the ancestry component. So this is more complex because ancestry and social demographic situations have a lot of confounding effects that may underlie the differences. But one thing that we also study in my lab is why women of African ancestry tend to get more of the higher grade endometrial cancer and die much more from these.
So there is a significant disparity here. And so one question that we ask, what is the endometrial epithelium makeup look like in African compared to other ancestries? And then we look at again, the master regulator transcription factor across these cell stages. And our hope is that we can understand things in the normal epithelium first, and then we can go and figure out how these things may underlie more aggressive cancer. So I don't have time to show you, but we have an expensive study on how endometrial cancer types are stratified across different demographics and what are the risk factors that can influence the disease severity. We study obesity and we study some of the chromatin regulators and epigenetic factors in that context, but using this normal dataset is going to also inform for us in that I would like to thank everyone in the lab, every collaborator, all the funding resources, but mostly you guys and specifically Dr. Kin and his team. It's been great to be coming to this foundation meeting every year and learning a lot and collaborating with most of you. I've been collaborating with OCI for some projects. Now I realize that we are collaborating with Dr. Taylor. And then of course Linda has been a long-term mentor and a collaborator. And I think that's it. But we will have a q and a, right? Okay, now it's q and a time. Perfect. Thank you so much for your attention.
So I stay here, right?
