From macrophages that seek out and destroy infectious agents to fibroblasts that hold tissues and organs together, cells give form and function to our bodies. However, despite their foundational roles in our biology, there is still much we don’t know about cells—like where different cell types are localized, what states a given cell type may take on, how the molecular characteristics of cells change over a person’s lifetime and more. Addressing these questions will provide a deeper understanding about the cellular and genetic basis of human health and disease.
One of the key efforts in this area is the Human Cell Atlas initiative. Like the Human Genome Project, the HCA is a consortium of researchers working to create a “map” of the diverse cell types found throughout the human body. Last month, researchers belonging to this initiative published four major studies that used RNA-sequencing to study single-cell gene expression across various tissue and cell types. Together, the studies provide a glimpse at the cellular diversity in the human body, offer new lines of inquiry for researching the cellular and genetic basis of diseases and demonstrate the power of modern sequencing and data analysis technologies.
Two of the papers provide an atlas of immune cell types, including a study on the location of immune cells in organs throughout the body and a study on how immune cells develop in various organs (1, 2). A third paper generated single-cell transcriptome analyses of multiple tissues collected from individual donors to examine how gene expression, mutations and splicing may vary in a single individual (3). The fourth paper uses single nucleus RNA sequencing (snRNA-seq) to create a molecular cross-organ reference map of cell types and cell states (4).
In this last paper, the researchers were able to use flash-frozen tissue samples from a tissue bank instead of fresh tissue samples from donors. Though all four papers are a tour de force of genomic and cellular biology, using frozen tissue bank samples to probe gene function in multiple cell types is a first and represents a particularly useful approach for future cell atlas efforts.
Creating a Cell Atlas with snRNA-seq
The benefit of using snRNA-seq is that it can be used to isolate and sequence RNA from cells where single cell RNA sequencing (scRNA-seq) wouldn’t be feasible, such as previously frozen cells and cells like fibroblasts that are resistant to dissociation. The technique, therefore, opens more cell and sample types to genetic analysis.
In this study, the researchers gathered frozen tissue samples of eight different tissue types from 16 different donors. The tissue types included breast, esophagus mucosa and muscularis, heart, lung, prostate, skeletal muscle and skin cells. They screened the tissue cells to confirm they had non-diseased pathology.
They then developed four different nucleus isolation methods to process the tissue samples for snRNA-sequencing, including one commercial isolation kit. After isolating the nuclei, they carried out droplet-based scRNA-seq. Using previously published reference genes and data analysis protocols that excluded contaminating non-nuclear RNA, the researchers were able to generate a total of 209,126 profiles of cellular nuclei with 918 genes detected for each profile, on average.
The researchers gathered previously reported cell type gene markers and used those markers to annotate the cells from their tissue samples into cell types and other subsets. Overall, they identified 43 broad cell classes, including cells that are difficult to analyze by scRNA-seq methods that require cells to dissociate prior to analysis, and could map those classes across the tissue types studied. Further analysis revealed previously unknown features of cell states, locations of certain cell types and potential associations between cell types and disease pathologies.
Check out the resources that Promega has to help you understand cell biology from reporter gene assays to understand gene expression to live-cell kinetic assays to follow cell biology in real time.
Five Key Findings
- In macrophage populations throughout the tissue samples, expression levels of LYVE1 and HLAII were dichotomous, that is, macrophage populations in tissues were either high in LYVE1 and low in HLAII expression or vice versa. Macrophages that are LYVE1high have previously been found to have potential regulatory roles, which was corroborated in this study. Conversely, HLAIIhigh cells were found to be more closely associated with immune processes.
- Lipid-associated macrophages (LAMs) have previously been reported in obese human adipose tissue, liver, lung and heart tissues as well as in leprosy and the brains of people with Alzheimer’s disease. In this study, the researchers found that LAM-like cells were associated with genes related to levels of cholesterol, trigylcerides and type 2 diabetes, confirming previous findings that these cells are involved with lipid regulation. They were able to identify transcription factors that were conserved across the identified LAM-like cells, which suggest that these cells share regulatory mechanisms, a potential target for drug development programs.
- Fibroblast-type cells were identified in lung tissue. These lung alveolar fibroblasts expressed mechanosensitive ion channels and genes related to actomyosin contraction, suggesting these cells are involved in sensing the expansion and contraction of lung tissues.
- In an analysis of the associations between cell types and muscle disease genes controlled by variation in a single gene, the researchers found that though many of the associations were conserved across human and mouse cell types, there were key differences. Most notably, expression of dystrophin (DMD) was different: in human cells, DMD expression was high in adipocytes in muscle tissue, whereas DMD expression in mouse muscle tissue adipocytes is low. Variations in DMD expression has been linked to Duchenne muscular dystrophy, and this finding suggests that DMD-linked disease states in mouse disease models may be different from human patients.
- One portion of the study examined links between disease-associated genes to specific tissues and cell types. One finding here was that loci associated with type 2 diabetes were enriched in skeletal muscle adipocytes and lymphatic endothelial cells across several tissue types. Individuals with type 2 diabetes are predisposed to vascular diseases, which may be reflected in the association of these loci and cell types.
This study revealed previously unknown locations, associations and expression profiles of a range of cell types, providing a wealth of new research directions and testable hypotheses related to cell biology and disease. The snRNA-seq method also means that the diverse tissue types and sources found in tissue banks are open to further cell atlas studies.
- Conde, C.D., et al. (2022) Cross-tissue immune cell analysis reveals tissue-specific features in humans. Science 376(6594) doi: 10.1126/science.abl5197
- Suo, C., et al. (2022) Mapping the developing human immune system across organs. Science 376(6597) doi: 10.1126/science.abo0510
- The Tabula Sapiens Consortium (2022) The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans. Science 376(6594) doi: 10.1126/science.abl4896
- Eraslan, G., et al. (2022) Single-nucleus cross-tissue molecular reference maps toward understanding disease gene function. Science 376(6594) doi: 10.1126/science.abl4290