Single-cell transcriptional profiling of human thymic stroma uncovers novel cellular heterogeneity in the thymic medulla

The thymus’ key function in the immune system is to provide the necessary environment for the development of diverse and self-tolerant T lymphocytes. While recent evidence suggests that the thymic stroma is comprised of more functionally distinct subpopulations than previously appreciated, the extent of this cellular heterogeneity in the human thymus is not well understood. Here we use single-cell RNA sequencing to comprehensively profile the human thymic stroma across multiple stages of life. Mesenchyme, pericytes and endothelial cells are identified as potential key regulators of thymic epithelial cell differentiation and thymocyte migration. In-depth analyses of epithelial cells reveal the presence of ionocytes as a medullary population, while the expression of tissue-specific antigens is mapped to different subsets of epithelial cells. This work thus provides important insight on how the diversity of thymic cells is established, and how this heterogeneity contributes to the induction of immune tolerance in humans.


Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences
Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative. For single-cell RNA sequencing, five donors were chosen to represent donor variability, which was sufficient to define broad cell populations representative of the human thymic stroma. The majority of our analyses had hundreds to thousands of cells for each cell type, which allowed identification of many statistically significant differentially expressed genes. In addition, we increased our sample size by merging our dataset with the publicly available dataset from the Park et al. study (DOI: 10.1126/science.aay3224). For immunofluorescence imaging, representative images from 4-5 fields from at least two donors were shown. For the lineage tracing experiment, no statistical methods were used to predetermine sample size. Sample size was chosen based on the magnitude and consistency of measurable differences between groups and application of standard practices within the field.
Computational cell and gene pre-filtering of single cell data was described in the Methods section. Cells with less than 200 or more than 5000 detected genes were excluded as well as cells expressing >10% mitochondrial genes.
Single-cell RNA sequencing was performed using cells from five donors. Merging our dataset with the Park et al. dataset containing cells from twelve donors largely replicated our findings. All other experimental findings described in this manuscript were also reliably reproduced between different donors.
Animals were age-matched and were allocated to groups based on genotype. Whenever possible, animals were co-housed with littermate controls and sex-matched.
Researchers were not blinded during experiments because our study results did not involve the combination of subjective decisions or a priori hypotheses, where classical blinding might be needed. Rather, we conducted computational analyses and genome-wide statistical tests, then reported the results of these agnostic analyses. Our findings were also supported by quantitative measure (i.e. flow cytometry) where blinding was unnecessary. Finally, investigators were not blinded to mouse genotypes in our experiments. This was to help ensure that appropriate sample size was achieved by sacrificing the minimum number of mice required to achieve consistent data. All primary antibodies were validated by the manufacturer and/or in previously published articles.
The specificity of primary antibodies for immunofluorescent staining was verified on human thymic sections using various dilutions (including the manufacturer's recommended dilution). Secondary antibody only controls were used to evaluate and confirm specificity of the antibodies to the respective epitopes. The following antibodies used for flow cytometry were similarly validated on human thymic preparations: EPCAM APC, BioLegend, 324208 CD45 Alx488, BioLegend, 304017 KRT8-Alx488, Abcam, ab192467 KRT5-Alx647 Abcam ab193895 KRT15-Alx555 Abcam ab214393 Rosa26CAG-stopflox-tdTomato and Ascl1-creERT2 mice were obtained from The Jackson Laboratory (RRID:IMSR_JAX:007914 and RRID:IMSR_JAX:012882). ADIG mice have been described previously (DOI: 10.1126/science.1159407). Both male and female mice aged 12-15 weeks were used. Mice were maintained at a constant humidity between 30-70% and temperature 68-79 degrees Fahrenheit, under a 12-h light/dark cycle and had free access to food and water.
The study did not involve wild animals.
Field samples were not collected for the study.

UCSF Institutional Animal Care and Use Committee (IACUC) and Laboratory Animal Resource Center
Thymic tissue donors were male and female between 19 weeks gestational age and 25 years old. Tissue was de-identified at the point of collection.