Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Aug 25;16(1):7680.
doi: 10.1038/s41467-025-62523-1.

An iPSC-based in vitro model recapitulates human thymic epithelial development and multi-lineage specification

Affiliations

An iPSC-based in vitro model recapitulates human thymic epithelial development and multi-lineage specification

Yann Pretemer et al. Nat Commun. .

Abstract

Thymic epithelial cells (TEC) are crucial in supporting T cell development, but their high heterogeneity and difficulty of isolation pose obstacles to their study in humans. Particularly, how diverse TEC lineages arise from a common progenitor remains poorly understood. To address this, here we establish a human iPSC-based model of thymus organogenesis capable of deriving these lineages in vitro. Through controlled retinoid signaling followed by self-directed differentiation, we obtain FOXN1+ TEC progenitor-like cells and diverse mature MHCII+ populations resembling cortical and medullary TECs, allowing us to infer their developmental trajectories. Upon thymocyte co-culture, induced TECs support the generation of naïve T cells with diverse TCR repertoires and further develop into AIRE+ and mimetic TEC subpopulations. Our system provides a fully in vitro model of human TEC differentiation from early fate specification to late-stage maturation, offering new insights into human thymus development and potential regenerative applications for congenital thymic disorders.

PubMed Disclaimer

Conflict of interest statement

Competing interests: Y.P., Y.G., and Y.H. are named inventors of patent application no. PCT/JP2023/021888, related to the induction method of thymic epithelial cells, filed by Kyoto University. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. TEC induction from human iPSCs in a 2D in vitro system.
a Schematic of the TEC induction protocol. D, day; iPSC, induced pluripotent stem cell; PS, primitive streak; DE, definitive endoderm; AFE, anterior foregut endoderm; PE, pharyngeal endoderm; TEP, thymic epithelial progenitor; TEC, thymic epithelial cell. b Phase-contrast images of induced cells at each time point. Scale bars, 200 µm. c Representative flow cytometry results (bottom) with the respective isotype controls (top) on D3 and D7. d mRNA expression of pluripotency, DE, and AFE markers over time. e Representative images of immunostaining on D18. Scale bars, 100 µm. f mRNA expression of PE and third pharyngeal pouch (3rd PP) markers over time. g Heatmap of the mean log2 fold change in mRNA expression in response to retinoic acid (RA) addition from D7 to D18 compared to DMSO only. h mRNA expression with or without addition of 50 ng/ml FGF8 from D7 to D18. i Heatmap of the mean log2 fold change in mRNA expression in response to addition of inhibitors or activators of specific signaling pathways from D18 to D28 compared to untreated (-) controls on D28. Black fields represent cases where expression was undetected in all experiments. The signaling modulators and concentrations are as follows: WNT inhibition, 0.2 µM, 1 µM, 5 µM XAV939; WNT activation, 0.4 µM, 2 µM, 10 µM CHIR99021; HH inhibition, 0.2 µM, 1 µM, 5 µM Cyclopamine; HH activation, 20 nM, 100 nM, 500 nM SAG; BMP inhibition, 40 nM, 200 nM, 1 µM LDN-193189; BMP activation, 4 ng/ml, 20 ng/ml, 100 ng/ml BMP4; NOTCH inhibition, 0.1 µM, 0.5 µM, 2.5 µM RO4929097; NOTCH activation, 0.2 mM, 1 mM, 5 mM Valproic acid; FGF inhibition, 40 nM, 200 nM, 1 µM BGJ398; FGF activation, 2 ng/ml, 10 ng/ml, 50 ng/ml FGF8. All results are from n = 3 independent experiments using 201B7 and values indicate the mean ± SEM (standard error of the mean). Statistical significance was determined using unpaired two-sided t-tests (n.s. no significant difference, *p < 0.05). Exact p-values in (h) are as follows: TBX1, 0.0106; PAX9, 0.0146; FOXN1, 0.0374.
Fig. 2
Fig. 2. Differentiation of cTEC- and mTEC-like cells over time.
a FOXN1mCherry fluorescence (left) and phase-contrast (right) images over time. Scale bars, 50 µm. D, day. Representative of n = 6 (D28 and D57) or n = 4 (D133) independent experiments. b Representative flow cytometry results of the reporter compared to the parent line at each time point. c Quantification of the percentage of mCherryhigh and mCherrylow cells using flow cytometry. All values indicate the mean ± SEM from n = 6 (D0 to D80) or n = 4 (D133) independent experiments. d, e mRNA expression of thymic epithelial cell (TEC) markers in the mCherryhigh, mCherrylow, and mCherry- (neg) populations over time. All values indicate the mean ± SEM from n = 6 (D0 to D80) or n = 3 (D133) independent experiments. cTEC, cortical TEC; mTEC, medullary TEC. f Representative images of immunostaining on D80 from n = 6 independent experiments. Scale bars, 50 µm. g Representative images of immunostaining on D133 from n = 3 independent experiments. Scale bars, 200 µm (top row) and 50 µm (middle row). The bottom row schematically depicts the PSMB11+ area (green), the KRT5+ area (magenta), and the overlapping junctional area (gray) as seen in the immunostaining. c, cTEC-like; m, mTEC-like; j, junction. All results were obtained using the 201B7 FOXN1mCherry reporter iPSC line.
Fig. 3
Fig. 3. iTECs express functional and mature TEC markers.
a mRNA expression in the mCherryhigh, mCherrylow, and mCherry- (neg) populations over time. D, day. b Representative images of immunostaining on D80 from n = 6 independent experiments. Scale bars, 50 µm (left three images) and 500 µm (right image). ce Representative flow cytometry results on D133 (c) with quantification of CD205+ cells in each of the EPCAM+-gated mCherryhigh, mCherrylow, and mCherry- populations (d) and quantification of HLA-DQ+, HLA-DR+, HLA-ABC+, and CD90+ cells in the mCherryhighCD205+, mCherryhighCD205-, mCherrylow, and mCherry- populations (e). In (c), colored gates correspond to the populations used for quantification in (e). All results were obtained using the 201B7 FOXN1mCherry reporter iPSC line. All values indicate the mean ± SEM from n = 6 (D0 to D80) or n = 3 (D133) independent experiments. TEC, thymic epithelial cell; iTEC, induced TEC.
Fig. 4
Fig. 4. Efficient in vitro production of CD4+ and CD8+ T cells with diverse TCR repertoires through iTEC co-culture.
a Schematic of induced thymic epithelial cell (iTEC) co-culture with double-positive (DP) thymocytes. SP, single-positive. b Phase-contrast image of an iTEC/thymocyte organoid during co-culture, representative of n = 4 independent experiments. Scale bar, 500 µm. c Representative flow cytometry results after 14 days of mCherry+ iTEC or MS5-hDLL1 organoid co-culture with DP thymocytes. d Quantification of co-culture results from (c), with values indicating the mean ± SEM from n = 4 independent experiments. Naïve phenotype is defined as CD62L+CD45RA+CCR7+ triple-positive. Statistical significance was determined using unpaired two-sided t-tests (n.s. no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). Exact p-values are as follows: TCRαβ+, 0.0000557; TCRγδ+, 0.0102; CD4+ SP, 0.00378; CD8+ SP, 0.0447; CD4+CD8+ DP, 0.00000877; CD8αβ+, 0.0257; naïve phenotype (in CD4+ SP), 0.000944; naïve phenotype (in CD8+ SP), 0.00123. e Gene usage of TRBV and TRBJ in CD4+ (top) and CD8+ (bottom) SP thymocytes obtained after 14 days co-culture of mCherry+ iTECs and DP thymocytes in n = 1 independent experiment. All iTEC results were obtained using the 201B7 FOXN1mCherry reporter iPSC line.
Fig. 5
Fig. 5. Single cell profiling of iTECs reveals heterogeneous TEC-like populations.
a, b Induced cells from D7 to D133 according to time point (a) and unsupervised clustering (b) on UMAP. D, day; AFE, anterior foregut endoderm; PE, pharyngeal endoderm; TEP, thymic epithelial progenitor; TEC, thymic epithelial cell; cTEC, cortical TEC; mTEC, medullary TEC; KC, keratinized cell; Neu, neuron; Myo, myocyte; Mes, mesenchyme; Cil, ciliated. c Expression of AFE, PE, and TEC marker genes on UMAP. d Module scores of the top 20 cTEC and mTEC marker genes as defined by Park et al.. e Dot plot of TEC and other marker gene expression in induced AFE-derived lineages. % expr., percentage of cells in which the gene is expressed; avg. expr., average expression. f Expression of cTEC, mTEC, and mature TEC marker genes on UMAP. g Heatmap of the average expression of the top 20 differentially expressed marker genes in each AFE-derived terminal cluster. All results were obtained using the 201B7 FOXN1mCherry reporter iPSC line.
Fig. 6
Fig. 6. Induced cTEC- and mTEC-like cells resemble human primary TECs.
a UMAP of unsupervised clustering of primary thymic stroma. TEC, thymic epithelial cell; cTEC, cortical TEC; mTEC, medullary TEC; mTEC-mim, mimetic mTEC; mcTEC, intertypical TEC; Mes, mesenchyme; Endo, endothelial cell; VSMC, vascular smooth muscle cell; Lympho, lymphocyte. b Expression of TEC markers in primary thymic stroma on UMAP. c Dot plot of selected cluster and lineage marker gene expression in the TEC populations of primary thymic stroma. % expr., percentage of cells in which the gene is expressed; avg. expr., average expression. d, e UMAP of induced TEC, KC, and Mes clusters from day 133, with populations and colors corresponding to Fig. 5b, integrated with primary thymic stroma (black). The primary mTEC-CLDN3/4 cluster and the co-clustering induced day 133 cells highlighted in (d) are shown separately in (e). f Expression of secretory, keratinized, and AIRE+ mTEC marker genes in the primary mTEC-CLDN3/4 cluster on UMAP. g Expression of secretory cell markers in induced cells on UMAP. h Violin plot of regulon activity in each cluster represented by scaled AUC (Area Under the Curve) scores from SCENIC, showing the top 15 induced mTEC marker regulons also found in primary mTECs. All results were obtained using primary thymic stroma from n = 3 pediatric donors and the 201B7 FOXN1mCherry reporter iPSC line.
Fig. 7
Fig. 7. Single cell trajectory analysis of iTECs models TEC development and lineage divergence.
a Slingshot-identified lineage trajectories (left) and RNA velocity by scvelo (right) on UMAP. AFE, anterior foregut endoderm; PE, pharyngeal endoderm; TEP, thymic epithelial progenitor; TEC, thymic epithelial cell; cTEC, cortical TEC; mTEC, medullary TEC; KC, keratinized cell. b UMAP showing the trajectories of each lineage from Slingshot colored by pseudotime. c Smoothed gene expression of cTEC and mTEC markers along pseudotime in each lineage. d Clustered heatmap of the scaled smoothed gene expression of the top 60 differentially expressed genes along pseudotime using the patternTest in tradeSeq. e Expression of selected genes from (d) showing the earliest lineage divergence from PE on UMAP. f Schematic of TEC lineage divergence inferred from pseudotime analysis of induced TECs (iTEC), displaying the clusters with clear developmental paths and associated lineages in primary TECs. All results were obtained using the 201B7 FOXN1mCherry reporter iPSC line.
Fig. 8
Fig. 8. iTECs recapitulate AIRE+ and post-AIRE mTEC development in thymocyte co-culture.
a Representative flow cytometry results of iTEC/thymocyte organoids after 2 weeks of co-culture from n = 6 independent experiments. TEC, thymic epithelial cell; iTEC, induced TEC. b mRNA expression of AIRE at the iPSC stage (iPS) and in the mCherry+ (co_pos) and mCherry- (co_neg) populations after 2 weeks of iTEC/thymocyte co-culture. All values indicate the mean ± SEM from n = 6 independent experiments. Statistical significance was determined using unpaired two-sided t-tests (n.s. no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001). The exact p-value is 0.000390. c UMAP of unsupervised clustering of cells after iTEC/thymocyte co-culture for 2 weeks from n = 1 independent experiment, consisting of 14 pooled organoids from one induction using four thymocyte donors with 3-4 organoids each. cTEC, cortical TEC; mTEC, medullary TEC; mTEC-mim, mimetic mTEC; Mes, mesenchyme; Lympho, lymphocyte; Neu, neuron; Myo, myocyte. d UMAP of the co-cultured cells from (c) integrated with the primary thymic stroma from Fig. 6a, showing the clustering of the primary thymic stroma alone (top) and the overlay of the iTEC/thymocyte co-culture cells (bottom). e, f Expression of TEC and regulatory T cell (Treg) marker genes in all co-cultured cells (e) and cTEC, mTEC, and neuronal marker genes in the EPCAM+ clusters (f) on UMAP. Smaller areas containing FOXP3+IL2RA+ cells in (e) and the majority of the AIRE+ cells in (f) are also shown magnified. g RNA velocity of the EPCAM+ clusters of co-cultured cells by scvelo on UMAP. h UMAP showing the latent time inferred from RNA velocity (left) and the pseudotime of the mTEC-mim II and III lineages derived by Slingshot (middle and right). The black arrowhead indicates the area with the earliest latent time. i Smoothed gene expression of epithelial progenitor, AIRE+ mTEC, and early neuronal differentiation markers, as well as selected mTEC-mim II and mTEC-mim III cluster markers, along pseudotime in each lineage. All results were obtained using the 201B7 FOXN1mCherry reporter iPSC line co-cultured with primary double-positive (DP) thymocytes.

References

    1. Kadouri, N., Nevo, S., Goldfarb, Y. & Abramson, J. Thymic epithelial cell heterogeneity: TEC by TEC. Nat. Rev. Immunol.20, 239–253 (2020). - PubMed
    1. Abramson, J. & Anderson, G. Thymic epithelial cells. Annu. Rev. Immunol.35, 85–118 (2017). - PubMed
    1. Takahama, Y., Ohigashi, I., Baik, S. & Anderson, G. Generation of diversity in thymic epithelial cells. Nat. Rev. Immunol.17, 295–305 (2017). - PubMed
    1. Farley, A. M. et al. Dynamics of thymus organogenesis and colonization in early human development. Development140, 2015–2026 (2013). - PMC - PubMed
    1. Nehls, M. et al. Two genetically separable steps in the differentiation of thymic epithelium. Science272, 886–889 (1996). - PubMed

LinkOut - more resources