In this recorded session, the chair introduces Mariana (Marianna), a bioinformatics lab leader at the Brazilian National Cancer Institute, who presents her work on macrophage subpopulations in both adipose tissue and tumor microenvironments. She reviews the functional diversity of macrophages (M1, M2, and the more nuanced plasticity beyond this dichotomy) and highlights previous single‑cell RNA‑seq analyses of mouse white adipose tissue that identified five adipose‑tissue macrophage (ATM) subtypes. Building on this, her team, supported by the Chan Zuckerberg Initiative, plans a three‑phase study of lean and obese Brazilian donors (≈12 donors in phase 1, later expanded with bulk RNA‑seq) to generate high‑resolution single‑cell profiles of subcutaneous and visceral adipose tissue. She also shares preliminary results from an integrated analysis of >40,000 myeloid cells across tumor and normal samples, revealing ~20 distinct clusters, including an unpolarized M0‑like cluster associated with poor melanoma survival and a potential novel subtype lacking known signatures. Tissue‑specific enrichment (e.g., lung vs. skin) and resident versus recruited macrophage origins are discussed, with ongoing pseudo‑time analyses to probe lineage dynamics. The talk concludes by emphasizing the importance of detailed macrophage characterization for understanding obesity and cancer, the potential for new immunotherapeutic targets, and invites collaborators and post‑doctoral candidates. Audience questions address expected numbers of subtypes, handling Brazil’s genetic diversity, ancestry effects on gene expression, the nature of M0 cells, and whether phenotypic differences reflect temporal stages of monocyte‑derived macrophages.
Chapters
Thank you everyone for working on research to fight the global pandemic COVID-19. It is my pleasure to chair this session highlighting single-cell research in Latin America. It's an honor to be introducing the session speakers. Note there will be a brief question and answer just as before after each presentation. Please do use Slido to submit your questions during the talk and during the question and answer, and I'm looking forward to reading them. To our, our speakers. As our first speaker, I would love to welcome Mariana, who has a PhD in bioinformatics and completed her postdoctoral work at the Gene Center at LMU in Munich, Germany. She leads the bioinformatics and computational biology lab at the Brazilian National Cancer Institute, where she conducts research focused on the interaction between tumor microenvironment and malignant cells and is starting a collaboration on single-cell analysis and inflammation. Where she will be profiling macrophage subpopulations present in adipose cells. Welcome, Marianna. Thank you so much for starting our session.
Hi, good afternoon. So first of all, I would like to thank the organizers for this invitation. It's a pleasure to me to be here today and share a little bit of our recent work associated with macrophage subpopulations. Microphages are a multifunctional cell type. They are present in most tissues in vertebrates, and their phagocytic activity is important not only for host defense against infections but also for a variety of housekeeping functions, such as removal of apoptotic cells and remodeling of extracellular matrix. Besides their role in regulating tissue remodeling and repair, macrophages are also well recognized by orchestrating an inflammatory response through antigen presentation and cytokine production, recruiting and activating cells from the immune system. However, macrophages are also associated with an aberrant and persistent inflammatory response, and in this case, it can lead to tissue degeneration and a variety of diseases such as cancer and metabolic syndromes. To exert this vast range of functions, they must present different phenotypes,, which are mostly a combination of differentiation and polarization programs. The polarized phenotypes are induced by signals, leading to a stable phenotypic change through the induction of different gene expression programs. The literature has extensively characterized two main activation states. Of macrophages, which are the classically activated macrophages, named M1, and the alternatively activated macrophages, named M2. So, M1 phenotype is associated with antigen presentation and inflammation eliciting, while M2 is is more characterized by tissue repair induction and inflammation reduction. Many diseases have been associated with increased number and activation of either M1 or M2 macrophages, such as the metabolic syndromes. And regarding obesity, Early studies suggested that there is a differential polarization of adipose tissue macrophages, with M2-like macrophages predominant in lean adipose tissue, while M1-like macrophages are therefore more enriched in obese adipose tissues. However, this dichotomy has been questioned since recent studies show that the phenotypic plasticity of macrophages is far more complicated. Several phenotypes with different functions can coexist in the adipose tissue, contributing to the development of obesity. And as far as we know, this has not been extensively explored in the literature. In order to take a look at the macrophage subtypes present in the mutant adipose tissue, we have analyzed single-cell RNA-seq data available in the literature from epithelial white adipose tissue samples. After processing the data, we selected the clusters expressing ADGRI-1, a widely used monocyte-macrophage marker in mice. Macrophage subclusters were then identified and inspected regarding the other markers depicted here in this slide. We also correlated the identified subpopulations with signatures from the literature and in order to better characterize the macrophages and monocytes subpopulations. We observed that the 5 ATM subtypes reflect different gene expression profiles, probably having a different function in the adipose tissue. Based on this result, we concluded that a high-resolution profile of myeloid-derived cells associated with the adipose tissue is needed. So therefore, with the support of the Chan Zuckerberg Initiative and together with my two colleagues, colleagues from the University of Campinas, Marcelo Mori and Pedro Moraes Vieira, we sought to characterize the macrophage subpopulations in the adipose tissue to investigate their role in obesity. We will soon start to profile those macrophage subpopulations of lean and obese donors from both subcutaneous and visceral adipose tissue of Brazilian donors using single-cell RNA-seq analysis. Well, we are very excited to start this project, but while we don't have any direct results to show to you regarding this project, I would like to show what we have been done regarding microphage subpopulations in the tumor microenvironment, where they have been described to exert dual functions that can either promote tumor progression or tumor killing. In a recent work of our group, we observed that metastatic melanoma samples showing predominant unpolarized M0 signature were associated with a poor for overall survival. We then decided to investigate how this macrophage plasticity is also impacting different tumor types. So therefore, we started to integrate several single-cell RNA-seq data derived from both tumor and normal samples from different sites in the body. Using these markers depicted here in this slide, we were able to select the myeloid-derived cells and analyze these different subpopulations and also its impact in tumor progression. So far, we were able to integrate more than 40,000 cells and identify 20 clusters. When analyzing genes differentially expressed, we saw that the myeloid cell subpopulations present distinctive gene expression profiles. And as you can see here, each cluster comprises a different number of cells, with the smallest one comprising around 250 cells. Cells. The cells in each cluster also express a similar number of transcripts, with some exceptions for cluster 3, for example. With these results, we were able to identify tumor or normal tissue-specific myeloid cells, pointed here by those arrows. Also, subpopulations that were more predominant in a specific specific sites, such as lung, represented here by the green bars, or skin, represented here by the purple bars. When evaluating the subpopulations heterogeneity by tissue type, melanoma was the most diverse sample in terms of different cell subtypes, while the healthy skin tissue tissue showed only a few subpopulations. In order to better characterize each cluster, we did a correlation with known signatures. We have identified interesting subpopulations such as the cluster 2, which shows a high correlation with the unpolarized M0 macrophage phenotype that has been only characterized in vitro so far. Well, these clusters were predominantly found in the tumor samples, and it expressed some important and red-associated genes, genes associated with tumor progression, such as SPP1, ALSO Marco, and STSB, corroborating some of our previous results published this year regarding metastatic melanoma. We also found a cluster with no correlation with the evaluated signatures that expressed TBS-AB1, suggesting a potential new macrophage subtype, but new analyses are needed in order to better characterize these cells. Well, although very preliminary, We are very excited about our recent results. So in conclusion, I want to reinforce the important role of microfaits associated with host defense, tissue development, homeostasis, and repair, but also its crucial role in the activation of basal chronic inflammation, leading to the development of diseases as we saw. Such as cancer and obesity. There are plenty of macrophage phenotypes should be better investigated, and we believe that the high-resolution characterization of their unique plasticity can help the development of new immunotherapeutic strategies. So with that, I would like to thank my entire team and also my my collaborators, especially the ones highlighted here, who contributed directly to the results shown here. Also, I would like to thank the funding agencies. And last but not least, we are seeking for postdoc candidate to work in this project, so if you are interested, please email me to have more info on that. And I I'll be glad to take your questions. Thank you.
Thank you so much, Maryana. And thank you everyone who is sending questions in through Slido. Thank you. So as those questions are coming in, we do have one. They indicate a great talk there. How many additional subsets of macrophages do you expect to find when you start doing single-cell RNA-seq, M0 to M300?
Question mark.
Okay, thank you.
So great question. Actually, when we started this work, we have no idea of how many subtypes we could actually found. Actually, it was a surprise that we could found at least 20. Different subtypes. And I think the secret here is that when we enhance the number of data that we are evaluating, then we can discover new subpopulations. So, I believe that if we keep enhancing the amount of data, we can to discover much more subtypes.
Makes a lot of sense. So I was wondering a little bit about the ongoing, I guess, work on the high-resolution dataset. I know, at least from my perspective, Brazil has a really wonderful diversity of population. So how do you approach that when you're wanting to do this high-resolution Brazilian population overview.
Okay, so well, the way we designed the project, it's gonna be like at least 3 phases. So the first one, we want to at least determine these subpopulations. And because this kind of experiment, it's a little bit expensive, let's say this way, it's not possible to profile many, many people in this search. So what we want to do is to discover very, like, real reliable markers, and then in a second phase, stage, we can profile more people in our population in order to better tackle this ethnic issue, you know. So in the first phase, we are gonna profile around 12 donors, if I'm not wrong. We have some both male and female donors, but I believe that we are going to enhance this knowledge in the second stage when we are going to use bulk RNA-seq to better characterize their signatures.
Thank you. And there's another question that came up. I think it's related to this. You may have answered it partially. So we go ahead and knock this one out. This is really cool work.
Congratulations.
Just one question regarding the study in Brazilians you're starting. Do you anticipate differences in gene expression depending upon ancestry? And we are new to these studies, but very interested in the value of incorporating individuals from different places.
Yeah, this is a great question. We actually right now are relying on datasets that are already available in the literature. So for sure, we need to include datasets from all around the world and then try to tackle this issue. I think it's a great question and we have to think about that. But we haven't touched it in this subject so far, mainly because of the datasets we have right now. But it's paramount to think about this and also to better plan the future experiments.
Another one. Are these M0 macrophages in the tumor resident-like macrophages?
No, actually they are from, well, we need to characterize them better, but our guessing is that they are derived from myeloid cells. So they are recruited ones. And I don't know, my hypothesis is that there is something like hampering their polarization. So, but this is just a guessing, we need to study them more because they haven't been characterized in vivo so far.
Only in vitro. Thank you. And you had so many questions coming in, so much excitement around your work. So we'll have time for one last one. Do these different phenotypes simply represent different stages of monocyte-derived macrophage once they enter the tissue and the amount of time they were in the tissue?
Yeah, this is what we believe.
We are performing right now some pseudo-time analysis. I think this is going to also help us to answer this question because then we can kind of better understand the plasticity of the cells. But because we are also analyzing normal tissues, we are also discovering also resident macrophages. So there are both types of cells there. The recruited one and also the resident ones. But it's very nice to see that depends on the site we are analyzing, there is a more specific cell type.
Well, thank you, Mariana. I would encourage others that had questions to please reach out to Mariana. We'll have to move on to our next talk.
