Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Runx factors launch T cell and innate lymphoid programs via direct and gene network-based mechanisms

Nature Immunology volume 24, pages 1458–1472 (2023)Cite this article

Subjects

An Author Correction to this article was published on 27 November 2023

This article has been updated

Abstract

Runx factors are essential for lineage specification of various hematopoietic cells, including T lymphocytes. However, they regulate context-specific genes and occupy distinct genomic regions in different cell types. Here, we show that dynamic Runx binding shifts in mouse early T cell development are mostly not restricted by local chromatin state but regulated by Runx dosage and functional partners. Runx cofactors compete to recruit a limited pool of Runx factors in early T progenitor cells, and a modest increase in Runx protein availability at pre-commitment stages causes premature Runx occupancy at post-commitment binding sites. This increased Runx factor availability results in striking T cell lineage developmental acceleration by selectively activating T cell-identity and innate lymphoid cell programs. These programs are collectively regulated by Runx together with other, Runx-induced transcription factors that co-occupy Runx-target genes and propagate gene network changes.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Runx transcription factors readily shift DNA binding site choice at different stages of T cell development largely independent of chromatin state changes.
Fig. 2: A mild increase in Runx factor availability prevents PU.1-mediated Runx1 depletion and prematurely upregulates Bcl11b, TCF1 and GATA3 in phase 1.
Fig. 3: Single-cell transcriptomes reveal that Runx-level perturbation causes cells to take different developmental paths deviating from the normal trajectory.
Fig. 4: Runx levels control T cell development progression rate by activating selective gene network modules.
Fig. 5: Runx1 overexpression results in overall faster T cell lineage development from DN1 to DN4 stages in the mixed-chimeric artificial thymic organoid.
Fig. 6: A modest increase of Runx1 concentration results in premature occupancy of post-commitment-preferred sites and new sites.
Fig. 7: Runx transcription factors control a gene regulatory network by cooperating with other transcription factors.

Similar content being viewed by others

Data availability

All new genomic sequencing data have been deposited in Gene Expression Omnibus under accession numbers GSE218147 (C&R, ChIP–seq and ATAC-seq) and GSE218149 (scRNA-seq). The publicly available data utilized for analysis are presented in Supplementary Table 7. Publicly available data were utilized by downloading raw sequence read files using the Sequence Read Archive toolkit Fastq-dump (v.2.10.9). Source data are provided with this paper. All other data needed to evaluate the conclusions in the paper are present in the paper, the Extended Data or the Supplementary Information, or are available upon request.

Code availability

All code used for data analysis in this work is publicly available and listed in the Methods and Reporting Summary.

Change history

References

  1. Shin, B. et al. Runx1 and Runx3 drive progenitor to T-lineage transcriptome conversion in mouse T cell commitment via dynamic genomic site switching. Proc. Natl Acad. Sci. USA 118, e2019655118 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Guo, Y., Maillard, I., Chakraborti, S., Rothenberg, E. V. & Speck, N. A. Core binding factors are necessary for natural killer cell development and cooperate with Notch signaling during T-cell specification. Blood 112, 480–492 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Zhao, L. et al. CBFB-MYH11 hinders early T-cell development and induces massive cell death in the thymus. Blood 109, 3432–3440 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Taniuchi, I. et al. Differential requirements for Runx proteins in CD4 repression and epigenetic silencing during T lymphocyte development. Cell 111, 621–633 (2002).

    Article  CAS  PubMed  Google Scholar 

  5. Egawa, T., Tillman, R. E., Naoe, Y., Taniuchi, I. & Littman, D. R. The role of the Runx transcription factors in thymocyte differentiation and in homeostasis of naive T cells. J. Exp. Med. 204, 1945–1957 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Yzaguirre, A. D., de Bruijn, M. F. & Speck, N. A. The role of Runx1 in embryonic blood cell formation. Adv. Exp. Med. Biol. 962, 47–64 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Growney, J. D. et al. Loss of Runx1 perturbs adult hematopoiesis and is associated with a myeloproliferative phenotype. Blood 106, 494–504 (2005).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Ichikawa, M. et al. AML-1 is required for megakaryocytic maturation and lymphocytic differentiation, but not for maintenance of hematopoietic stem cells in adult hematopoiesis. Nat. Med. 10, 299–304 (2004).

    Article  CAS  PubMed  Google Scholar 

  9. Talebian, L. et al. T-lymphoid, megakaryocyte, and granulocyte development are sensitive to decreases in CBFβ dosage. Blood 109, 11–21 (2007).

  10. Seo, W., Ikawa, T., Kawamoto, H. & Taniuchi, I. Runx1-Cbfβ facilitates early B lymphocyte development by regulating expression of Ebf1. J. Exp. Med. 209, 1255–1262 (2012).

  11. Niebuhr, B. et al. Runx1 is essential at two stages of early murine B-cell development. Blood 122, 413–423 (2013).

    Article  CAS  PubMed  Google Scholar 

  12. Pham, T.-H. et al. Mechanisms of in vivo binding site selection of the hematopoietic master transcription factor PU.1. Nucleic Acids Res. 41, 6391–6402 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Ungerback, J. et al. Pioneering, chromatin remodeling, and epigenetic constraint in early T-cell gene regulation by SPI1 (PU.1). Genome Res. 28, 1508–1519 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  14. Pencovich, N. et al. Cell-autonomous function of Runx1 transcriptionally regulates mouse megakaryocytic maturation. PLoS ONE 8, e64248 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Zang, C., Luyten, A., Chen, J., Liu, X. S. & Shivdasani, R. A. NF-E2, FLI1 and RUNX1 collaborate at areas of dynamic chromatin to activate transcription in mature mouse megakaryocytes. Sci. Rep. 6, 30255 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Miyazaki, M. et al. The opposing roles of the transcription factor E2A and its antagonist Id3 that orchestrate and enforce the naive fate of T cells. Nat. Immunol. 12, 992–1001 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Lin, Y. C. et al. A global network of transcription factors, involving E2A, EBF1 and Foxo1, that orchestrates B cell fate. Nat. Immunol. 11, 635–643 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zhong, Y. et al. Hierarchical regulation of the resting and activated T cell epigenome by major transcription factor families. Nat. Immunol. 23, 122–134 (2022).

    Article  CAS  PubMed  Google Scholar 

  19. Hosokawa, H. et al. Bcl11b sets pro-T cell fate by site-specific cofactor recruitment and by repressing Id2 and Zbtb16. Nat. Immunol. 19, 1427–1440 (2018).

  20. Hosokawa, H. & Rothenberg, E. V. How transcription factors drive choice of the T cell fate. Nat. Rev. Immunol. 21, 162–176 (2021).

    Article  CAS  PubMed  Google Scholar 

  21. Hu, G. et al. Transformation of accessible chromatin and 3D nucleome underlies lineage commitment of early T cells. Immunity 48, 227–242 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Yoshida, H. et al. The cis-regulatory atlas of the mouse immune system. Cell 176, 897–912 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Zhang, J. A., Mortazavi, A., Williams, B. A., Wold, B. J. & Rothenberg, E. V. Dynamic transformations of genome-wide epigenetic marking and transcriptional control establish T cell identity. Cell 149, 467–482 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Bruno, L. et al. Selective deployment of transcription factor paralogs with submaximal strength facilitates gene regulation in the immune system. Nat. Immunol. 20, 1372–1380 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Kueh, H. Y. et al. Asynchronous combinatorial action of four regulatory factors activates Bcl11b for T cell commitment. Nat. Immunol. 17, 956–965 (2016).

  26. Hamey, F. K. et al. Reconstructing blood stem cell regulatory network models from single-cell molecular profiles. Proc. Natl Acad. Sci. USA 114, 5822–5829 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Kitagawa, Y. et al. Guidance of regulatory T cell development by Satb1-dependent super-enhancer establishment. Nat. Immunol. 18, 173–183 (2017).

    Article  CAS  PubMed  Google Scholar 

  28. Wilson, N. K. et al. Combinatorial transcriptional control in blood stem/progenitor cells: genome-wide analysis of ten major transcriptional regulators. Cell Stem Cell 7, 532–544 (2010).

    Article  CAS  PubMed  Google Scholar 

  29. Okuyama, K. et al. PAX5 is part of a functional transcription factor network targeted in lymphoid leukemia. PLoS Genet. 15, e1008280 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Hosokawa, H. et al. Transcription factor PU.1 represses and activates gene expression in early T cells by redirecting partner transcription factor binding. Immunity 48, 1119–1134 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Skene, P. J., Henikoff, J. G. & Henikoff, S. Targeted in situ genome-wide profiling with high efficiency for low cell numbers. Nat. Protoc. 13, 1006–1019 (2018).

    Article  CAS  PubMed  Google Scholar 

  32. Lieberman-Aiden, E. et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289–293 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Ernst, J. & Kellis, M. Chromatin-state discovery and genome annotation with ChromHMM. Nat. Protoc. 12, 2478–2492 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Isoda, T. et al. Non-coding transcription instructs chromatin folding and compartmentalization to dictate enhancer–promoter communication and T cell fate. Cell 171, 103–119 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Bonifer, C., Levantini, E., Kouskoff, V. & Lacaud, G. Runx1 structure and function in blood cell development. Adv. Exp. Med. Biol. 962, 65–81 (2017).

    Article  CAS  PubMed  Google Scholar 

  36. Friedman, A. D. Cell cycle and developmental control of hematopoiesis by Runx1. J. Cell. Physiol. 219, 520–524 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Yu, M. et al. Direct recruitment of polycomb repressive complex 1 to chromatin by core binding transcription factors. Mol. Cell 45, 330–343 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Schmitt, T. M. & Zúñiga-Pflücker, J. C. Induction of T cell development from hematopoietic progenitor cells by delta-like-1 in vitro. Immunity 17, 749–756 (2002).

    Article  CAS  PubMed  Google Scholar 

  39. Wong, W. F. et al. Over-expression of Runx1 transcription factor impairs the development of thymocytes from the double-negative to double-positive stages. Immunology 130, 243–253 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Hosokawa, H. et al. Stage-specific action of Runx1 and GATA3 controls silencing of PU.1 expression in mouse pro-T cells. J. Exp. Med. 218, e20202648 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Hosokawa, H. et al. Cell type-specific actions of Bcl11b in early T-lineage and group 2 innate lymphoid cells. J. Exp. Med. 217, e20190972 (2020).

    Article  PubMed  Google Scholar 

  42. Miyazaki, M. et al. The E-Id protein axis specifies adaptive lymphoid cell identity and suppresses thymic innate lymphoid cell development. Immunity 46, 818–834 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Zhou, W., Gao, F., Romero-Wolf, M., Jo, S. & Rothenberg, E. V. Single-cell deletion analyses show control of pro-T cell developmental speed and pathways by Tcf7, Spi1, Gata3, Bcl11a, Erg and Bcl11b. Sci. Immunol. 7, eabm1920 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Wiede, F. et al. PTPN2 regulates T cell lineage commitment and αβ versus γδ specification. J. Exp. Med. 214, 2733–2758 (2017).

  45. Yao, Z. et al. Stat5a/b are essential for normal lymphoid development and differentiation. Proc. Natl Acad. Sci. USA 103, 1000–1005 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Seet, C. S. et al. Generation of mature T cells from human hematopoietic stem and progenitor cells in artificial thymic organoids. Nat. Methods 14, 521–530 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Montel-Hagen, A. et al. In vitro recapitulation of murine thymopoiesis from single hematopoietic stem cells. Cell Rep. 33, 108320 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Van de Sande, B. et al. A scalable SCENIC workflow for single-cell gene regulatory network analysis. Nat. Protoc. 15, 2247–2276 (2020).

    Article  PubMed  Google Scholar 

  49. Brewster, R. C. et al. The transcription factor titration effect dictates level of gene expression. Cell 156, 1312–1323 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Lie-A-Ling, M. et al. Regulation of RUNX1 dosage is crucial for efficient blood formation from hemogenic endothelium. Development 145, dev149419 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  51. Lacaud, G., Kouskoff, V., Trumble, A., Schwantz, S. & Keller, G. Haploinsufficiency of Runx1 results in the acceleration of mesodermal development and hemangioblast specification upon in vitro differentiation of ES cells. Blood 103, 886–889 (2004).

    Article  CAS  PubMed  Google Scholar 

  52. Ng, K. K. et al. A stochastic epigenetic switch controls the dynamics of T-cell lineage commitment. eLife 7, e37851 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  53. Dionne, C. J. et al. Subversion of T lineage commitment by PU.1 in a clonal cell line system. Dev. Biol. 280, 448–466 (2005).

    Article  CAS  PubMed  Google Scholar 

  54. Telfer, J. C., Hedblom, E. E., Anderson, M. K., Laurent, M. N. & Rothenberg, E. V. Localization of the domains in Runx transcription factors required for the repression of CD4 in thymocytes. J. Immunol. 172, 4359–4370 (2004).

    Article  CAS  PubMed  Google Scholar 

  55. Meers, M. P., Bryson, T. D., Henikoff, J. G. & Henikoff, S. Improved CUT&RUN chromatin profiling tools. eLife 8, e46314 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

  56. Skene, P. J. & Henikoff, S. An efficient targeted nuclease strategy for high-resolution mapping of DNA binding sites. eLife 6, e21856 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

  57. Levanon, D. et al. Transcription factor Runx3 regulates interleukin-15-dependent natural killer cell activation. Mol. Cell. Biol. 34, 1158–1169 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  58. Buenrostro, J. D., Wu, B., Chang, H. Y. & Greenleaf, W. J. ATAC-seq: a method for assaying chromatin accessibility genome-wide. Curr. Protoc. Mol. Biol. 109, 21.29.1–21.29.9 (2015).

    PubMed  Google Scholar 

  59. Corces, M. R. et al. An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat. Methods 14, 959–962 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. & Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213–1218 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Liu, N. et al. Direct promoter repression by BCL11A controls the fetal to adult hemoglobin switch. Cell 173, 430–442 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. McLean, C. Y. et al. GREAT improves functional interpretation of cis-regulatory regions. Nat. Biotechnol. 28, 495–501 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Ramirez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W160–W165 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  66. Ernst, J. & Kellis, M. ChromHMM: automating chromatin-state discovery and characterization. Nat. Methods 9, 215–216 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  67. Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  68. Qiu, X. et al. Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods 14, 979–982 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Trapnell, C. et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat. Biotechnol. 32, 381–386 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  70. Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat. Methods 14, 1083–1086 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Romero-Wolf, M. et al. Notch2 complements Notch1 to mediate inductive signaling that initiates early T cell development. J. Cell Biol. 219, e202005093 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank the E.V.R. laboratory members for helpful discussions, R. Diamond and members of the Caltech Flow Cytometry and Cell Sorting facility for sorting, I. Antoshechkin and V. Kumar of the Caltech Jacobs Genomics Facility for sequencing, H. Amrhein and D. Trout for computer support, J. Park and S. Chen from the Caltech Single Cell Profiling and Engineering Center for support for processing 10x Chromium samples, J. Vielmetter and the Caltech Protein Expression Center for purifying protein A-MNase, I. Soto for mouse care, M. Quiloan and M. Chau for mouse genotyping and supervision, and J. Longmate (formerly City of Hope) and M. Yui (formerly Caltech) for statistics advice. Support for this project came from USPHS grants (R01AI135200, R01HL119102 and R01HD100039) to E.V.R., and from a Cancer Research Institute Irvington Postdoctoral Fellowship CRI.SHIN and Caltech Baxter Fellowship (to B.S.). F.G. was supported in part by the National Institutes of Health award 1RF1NS122060-01. Support also came from The Beckman Institute at Caltech for all the Caltech facilities and from the Biology and Biological Engineering Division Bowes Leadership Chair Fund, the Louis A. Garfinkle Memorial Laboratory Fund and the Al Sherman Foundation to the E.V.R. laboratory. E.V.R. gratefully acknowledges support from the Edward B. Lewis Professorship and past support from the Albert Billings Ruddock Professorship.

Author information

Author notes

  1. Wen Zhou

    Present address: BillionToOne, Menlo Park, CA, USA

  2. Fan Gao

    Present address: Lyterian Therapeutics, South San Francisco, CA, USA

Authors and Affiliations

  1. Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, CA, USA

    Boyoung Shin, Wen Zhou, Jue Wang, Fan Gao & Ellen V. Rothenberg

  2. Program in Biochemistry and Molecular Biophysics, California Institute of Technology, Pasadena, CA, USA

    Wen Zhou & Jue Wang

  3. Bioinformatics Resource Center, Beckman Institute of California Institute of Technology, Pasadena, CA, USA

    Fan Gao

Authors
  1. Boyoung Shin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wen Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jue Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Fan Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Ellen V. Rothenberg
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

B.S. and E.V.R. conceptualized the project, wrote the paper and edited the paper. B.S. performed the experiments, and analyzed data with W.Z. and J.W. F.G. wrote the in-house bioinformatic pipeline for hashtag alignment and provided further analysis. E.V.R. supervised research, acquired funding and provided additional data analysis. All authors edited the paper and provided helpful comments.

Corresponding author

Correspondence to Ellen V. Rothenberg.

Ethics declarations

Competing interests

W.Z. is employed by BillionToOne and has been employed by 10x Genomics (CA 94588). F.G. is employed by Lyterian Therapeutics. E.V.R. was a member of the Scientific Advisory Board for Century Therapeutics and has advised Kite Pharma and A2 Biotherapeutics. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Immunology thanks Artem Barski, Golnaz Vahedi and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. L. A. Dempsey was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Distinct motif enrichment patterns in dynamically shifting Runx binding sites, comparison with PU.1 site stability, and efficient detection of direct Runx binding sites by CUT&RUN.

a, Heatmap illustrates PU.1 binding profiles in immortalized HSPC, DN1, DN2a, and DN2b cells. b, Top motifs enriched in Runx binding sites from indicated regions of Fig. 1b are shown. Statistical significance was computed using Homer de novo motif discovery algorithm. c, Scatter plots and Area-proportional Venn diagrams compare Runx1 and Runx3 binding sites in Phase 1 and Phase 2 pro-T cells, as measured by ChIP-seq cross-linked with DSG + FA1 vs. by CUT&RUN (C&R). Numbers in the Venn diagram indicate number of differential peaks compared between DSG-crosslinked ChIP-seq vs. C&R (fold enrichment > 2, Poisson enrichment P < 0.001). d, Violin plots show Runx motif quality position weight matrix (PWM) score in non-promoter Runx peaks detected similarly by ChIP-seq and C&R (purple) or preferentially detected by different technique (green; more efficiently detected by C&R, red; more efficiently detected by ChIP-seq). The horizontal dotted black line shows threshold PWM score to be recognized as a Runx motif. Thin vertical black bars mark minima and maxima values and thick vertical black lines indicate 25th to 75th percentiles range. The red lines with white circles show median values. e, Motif frequencies for Runx, bHLH, ETS, and PU.1 or TCF1 (Tcf7) factors within each peak are displayed.

Source data

Extended Data Fig. 2 Runx TFs predominantly interact with active large-scale chromatin compartments, yet local chromatin state is not a major barrier for stage-specific redeployment of Runx factors.

a, Percentages of ChIP-seq or C&R-detected Runx binding sites that are open or closed at a given stage is shown. Only non-promoter sites were calculated as most of the promoter sites are stably accessible. b, Bar graph displays percentages of Runx binding sites detected by ChIP-seq1 or C&R that are in compartment A or compartment B regions. c, Bar graph depicts the percentage of ATAC-defined open chromatin regions among Runx binding sites within compartment (Comp) A, B, or N. d, Genomic regions were assigned to compartment A (active, HiC PC1 value ≥ 10), compartment B (inactive, HiC PC1 value ≤ −10), and compartment N (neutral, −10 < HiC PC1 value < 10) in 1 kb-bins from DN1 (ETP), DN2, and DN3 cells (data from ref. 21). Regions stably maintaining compartment states (A-to-A or B-to-B) vs. the regions undergoing compartment flipping were categorized and their enrichments within different groups of Runx binding sites or total genomic regions were compared. Graphs in inset show expanded-scale view from Extended Data Fig. 2d to record rare changes in genomic compartment association during DN1 (ETP) to DN3 progression. e, Representative UCSC genome browser tracks near Bcl11b and Ets1 regions show compartment state (represented with HiC PC1 values), DN1 and DN3 Runx occupancies, and published ATAC signals with CTCF and SMC3 ChIP-seq. f, Heatmaps represent distinct chromatin states computed using ChromHMM. The enrichments of different histone marks, ATAC, and loop-forming machineries (CTCF, SMC3) with each chromatin state are shown in purple (left). Genomic annotation for chromatin states is displayed in orange (middle). Enrichment with different groups of Runx peaks is illustrated in blue (right). Constitutively occupied Group 3 peaks and promoter peaks were enriched among constitutively active chromatin states (states 8 and 9), as expected, and Group 1 peaks (losing Runx binding from Phase 1 to Phase 2) had the highest enrichment within Phase 1-preferential active states (states 1 and 2). In contrast, the Group 2 sites newly occupied during commitment were more enriched among constantly accessible regions with weak H3K4me2 marks (state 10), even more than they were enriched for Phase 2-specific active states (states 5 and 6). Constitutively active states (states 8 and 9) were also enriched for Group 2 peaks, and Group 2 peaks were also the only group showing enrichment among sites that were largely ATAC-closed in both Phase 1 and Phase 2 (state 7). g, UCSC genome browser profile shows Runx factor binding patterns near Meis1 locus in Phase 1 and Phase 2, together with ATAC-seq and histone ChIP-seq (H3K4me2, H3K27me3, CTCF, SMC3) tracks. ChromHMM chromatin states are displayed as a colormap at the bottom.

Source data

Extended Data Fig. 3 Increase in Runx1 protein availability changes Runx1 binding site choices in the presence of PU.1 in DN3-like cells.

a, Histograms show protein expression levels of PU.1 and Runx1 after introducing PU.1 and/or Runx1-expressing vectors. Numbers within histograms indicate geometric mean fluorescent intensities (gMFI). Bar graphs summarize gMFIs of PU.1 and Runx1 with means and standard deviations. 6 independent experiments. Mann-Whitney test. For PU.1 gMFI comparisons, *** P = 0.0006. For Runx1 gMFI comparisons, *** P = 0.0006 and ** P = 0.007. b, Expression of non-T-lineage markers, CD11b and CD11c, were measured using flow cytometry. Bar graphs show frequencies of cells that do not express these markers. Mean and standard deviation from 6 independent experiments are displayed. Two-way ANOVA with Šídák’s multiple comparisons. *** P < 0.0001. c, Density plots display motif frequencies for Runx1 and PU.1 in each peak and violin plots illustrate the best motif qualities for Runx1 and PU.1 in a given peak. The horizontal dotted black line shows threshold PWM score to be recognized as a Runx or PU.1 motif. Thin black lines mark minima and maxima values and thick vertical black bars show 25th to 75th percentiles range. The red lines with white circles indicate median values. Two sample Kolmogorov-Smirnov (KS) tests, comparing each to motif scores of ‘Common’ peaks defined in Fig. 2b. *** P < 0.0001. n of Common peaks = 4,047, n of PU.1-depleted peaks = 5,081, n of PU.1-induced peaks = 5,369, n of OE new peaks = 17,085.

Source data

Extended Data Fig. 4 Protein expression levels of developmentally important transcription factors are sensitive to Runx factor dosage in Phase 1.

a, Representative histograms display Runx1 (top) or Runx3 (bottom) expression levels at 2 days (left) or 4 days (right) after transducing Runx1 OE or empty-control vectors. Cells were gated on live alternative lineage- infection+ cells, then separated as cKithi CD25− (DN1) and cKithi CD25+ (DN2a) populations. Graph summarizes results from 4–8independent experiments. Mann-Whitney test. *** P = 0.0002 for Runx1 gMFI on d2pi (n = 8), *** P < 0.0006 and ** P = 0.007 for Runx1 gMFI on d4pi (n = 6). For comparing Runx3 gMFI, * P = 0.0411 for d2pi (n = 6) and * P = 0.0286 for d4pi (n = 4). b, Runx1, Runx2, and Runx3 mRNA expression levels in Runx1 OE, Runx1/Runx3 knock-out (KO), and control cells. Runx transcript levels were measured using single cell RNA-seq described in Fig. 3. Differentially expressed gene statistics were calculated using a Wilcoxon test employing a Seurat algorithm. Bounds of box extend from 25th to 75th percentiles, median is indicated using horizontal line, and whiskers show 1.5⨯ the interquartile range. Points show outliers. c, Flow cytometry analysis strategy to test Runx1 dosage effects on regulating other TFs. Runx1 expression levels in control vs. Runx1 OE cells were compared using histogram overlay, then cells expressing low or mid-levels of Runx1 (Runx1low/mid) and cells with high levels of Runx1 (Runx1hi) were determined. d, TCF1, GATA3, and PU.1 expression levels in Runx1low/mid vs. Runx1hi cells in control and Runx1 OE cells were compared. Numbers inside the histograms are gMFI values. Graphs summarize 6 experiments. Mann-Whitney test. * P = 0.04 for Runx3, ** P = 0.0079 and * P = 0.0321 for TCF1, * P = 0.04 for GATA3, ** P = 0.0079 and * P = 0.0159 for PU.1, ns = not significant.

Source data

Extended Data Fig. 5 Single-cell transcriptome analyses of Runx perturbations: deviations from normal developmental clusters due to effects on core target genes responding to both gain- and loss-of-function.

a-b, tSNE 1-2 (a) and UMAP1-2 (b) display transcriptomes of control- or Runx1 overexpressed (OE) or Runx1/Runx3 knockout (KO) cells at indicated timepoints (left). Genes associated with different stages of cell cycles are illustrated on tSNE 1-2 (right). Top panels show location of cells before cell-cycle regression and bottom panels illustrates distribution of cells after cell-cycle regression. Note that Runx1 OE tends to shift population toward G1 while KO shifts cells toward G2/M, but Runx perturbation states do not separate well on tSNE 1 or UMAP1 axis. c, Cluster distributions of indicated Runx-perturbation conditions are shown. Size of each dot represents number of cells and colormap indicates z-score from standard residual analysis followed by Fisher’s exact test. d, Expression patterns of stem or myeloid-associated genes, Cd81, Csf2b, Meis1, and Ifngr2 are displayed on UMAP2-3 axes. e, Scatter plots compare Log2 fold-changes (FC) in gene expression between Runx1 OE vs. control or Runx KO vs. control populations at different timepoints (d2 vs. d4 after introducing OE conditions, d3 vs. d5 after delivering gRNA for KO conditions). Each dot represents a different gene. D2pi OE DEGs vs. d4pi OE DEGs Pearson correlation r = 0.87 and d3pi KO DEGs vs. d6pi KO DEGs Pearson correlation r = 0.67. n of d2pi OE DEGs = 337, n of d4pi OE DEGs = 468, n of d3pi OE DEGs = 189, n of d6pi OE DEGs = 249. f, Heatmap illustrates expression profiles of the common Runx-target genes sensitively responding to both Runx1 OE and Runx KO. Each cluster is sorted by normal developmental progression order.

Extended Data Fig. 6 Runx1 overexpression inhibited myeloid and granulocyte program, while supporting NK cell program even after inducing Bcl11b expression.

a, Schematics illustrate experimental design for competitive commitment assay. Empty control or Runx1 overexpression vectors expressing different markers were cultured with OP9-Dll1 to initiate T cell development. After 2 days, DN1, Bcl11b- DN2a, and Bcl11b+ DN2a cells were each sorted from each condition. The same number (100 cells) of the same stage cells from control and Runx1 OE conditions were co-cultured with Notch-signaling (OP9-Dll1) or Notch non-signaling (OP9-Control) stromal cells for 6 days, supplemented with IL-7 and Flt3-ligand. Throughout Extended Data Fig. 6, data derived from the indicated input populations were from 2 independent experiments, totals of 8 wells for OP9-Dll1 and totals of 11 or 12 wells of each input population for OP9-Control. For OP9-Dll1, n of 4 wells were analyzed for each population in both Exp1 and Exp 2. For OP9-Control, n of 6 wells were used for DN1, Bcl11b− DN2a, Bcl11b+ (late) DN2a (Exp1), and n of 6 wells for DN1, Bcl11b− DN2a and 5 wells for Bcl11b+ (late) DN2a (Exp2). Data point shapes indicate independent experiments (circle for Exp1 or triangle for Exp2). b, Representative flow plots show competition outcomes between Control (x-axis) vs. Runx1 OE (y-axis) from each condition. Graphs summarize the absolute numbers and frequencies of vector-expressing control vs. Runx1 OE populations in both conditions. Runx1 OE cells were disfavored with and without Notch signaling. Wilcoxon signed-rank test. For OP9-Dll1 numbers, * P = 0.0078 for Bcl11b− DN2a and Bcl11b+ DN2a. For OP9-Control numbers, *** P = 0.0005 for DN1 and Bcl11b− DN2a, and ** P = 0.0098 for Bcl11b+ DN2a. For OP9-Dll1 frequencies, * P = 0.0078 for Bcl11b− DN2a and Bcl11b+ DN2a. For OP9-Control frequencies, *** P = 0.0005 for DN1 and Bcl11b− DN2a, and *** P = 0.001 for Bcl11b+ DN2a. ns=not significant. c, Expression of NK1.1 vs. Ly6G/Ly6C were measured by flow cytometry after culture without Notch signals. Graphs show frequencies of cells expressing Ly6G/Ly6C or NK1.1 in cells. Wilcoxon signed-rank test. *** P = 0.0005 for % Ly6G/Ly6C+cells, *** P = 0.0010 and ** P = 0.0034 for % NK1.1+ cells, ns=not significant.

Source data

Extended Data Fig. 7 Elevated Runx1 levels in Phase 1 resulted in additional Runx occupancies in post-commitment preferred sites and closed chromatin regions.

a, Gating strategy to sort control and Runx1 OE Phase 1 cells for C&R is illustrated. Briefly, bone-marrow progenitor cells were co-cultured with OP9-Dll1 cells for 2 days, and empty control or Runx1 overexpressing vector was retrovirally introduced. After 40–42 hours (total 4 days of culture on OP9-Dll1 cells), infection+ Phase 1 cells were sorted. In this system, for most cells to reach Phase 2 normally, 8–10 days of culture are needed1,71]. b, Scatter plots and Venn diagrams compare differential Runx1 occupancies at promoter (top) and non-promoter regions (bottom) when Runx1 concentration was increased. Numbers indicate differential Runx1 binding sites (fold enrichment > 2, Poisson enrichment P < 0.001). c, Runx1 C&R signal intensities from indicated cells are shown. Note increased occupancy even at Group 1 and Group 3 sites which were already bound in control Phase 1 cells. d, Bar graph represents compartment state profiles within different groups of Runx binding sites. e, Testing hypothesis that Runx1 OE accesses sites conditionally occupied in other pro-T related contexts. Area-proportional Venn diagrams show analysis strategy to identify Runx binding sites appearing specifically in Bcl11b knockout DN2b/DN3 cells (left), and ILC2-specific Runx binding sites (middle; Runx1, right; Runx3). f, Bar graph shows percentages of Group 4 peaks overlapping with indicated Runx binding site types. g, Density plots illustrate motif frequencies for PU.1, TCF1 (Tcf7), bHLH, and GATA factors in different types of Runx binding sites. h, Runx1, Runx3 (blue), PU1 (purple)23, TCF1 (red), E2A and HEB (green) binding profiles in non-promoter regions under unperturbed Phase 1 or Phase 2 conditions are shown. Runx1 binding patterns in empty vector control and Runx1 OE transduced conditions are displayed in orange tracks (left). Stage-preferential dynamic binding groups are indicated as color bars. Group 1, Phase 1-preferential; Group 2a, Phase 2-preferential and precociously occupied by OE; Group 2b, Phase 2-preferential but not occupied by OE; Group 3, Phase 1 & Phase 2 shared; Group 4op, OE-specific and open sites; Group 4cl, OE-specific and closed sites. The numbers on the right side indicate percent of each group of peaks within the same color bar. TCF1, E2A, and HEB binding sites were measured in independent replicates using C&R from thymic DN3 cells. PU.1 occupancy was previously determined using ChIP-seq23.

Source data

Extended Data Fig. 8 Runx factors engage functional target gene regions together with PU.1, TCF1 and E-proteins.

UCSC genome browser tracks show Runx binding patterns (orange tracks, experimental conditions; blue tracks, unperturbed pro-T cells), PU.1 in DN1 cells, TCF1 in DN3 cells, E2A and HEB in DN3 cells, and ATAC-seq signals (black) in Phase 1 (DN1) and Phase 2 (DN2b) cells. a, Regulatory regions for genes highly expressed in Phase 1 (Plek, Lmo2, and Meis1) are displayed. b-c, Regulatory regions for genes highly expressed in Phase 2 are shown. b, Light red highlights mark Group 2a sites near Cd3 clusters, Tcf7, and Thy1 occupied under conditions of Runx1 OE. c, Genomic regions near Gata3 and Myb show indicated TF binding profiles. Green highlights mark co-occupancy of Runx, PU.1, TCF1, and E-proteins.

Extended Data Fig. 9 Distinct associations of transcriptional regulatory function with different groups of Runx binding sites.

a, Specific associations between different classes of Runx binding sites and Runx DEGs are tested using Fisher’s exact test (all DEGs vs. non-DEGs). Graphs visualize the percentages of the genes that have at least one peak of the indicated group (height of the spike) and the absolute number of genes possessing at least one such Runx binding peak, in the various DEG types (size of hexagon). Gray bars to the left of each plot indicate the percentages of genes associated with each peak type among Runx non-responding genes (non-DEGs), and broken line uses this level as a reference for DEG enrichment. All site types except Group 4cl sites were significantly enriched among DEGs relative to non-DEGs (shown by relatively higher of spike heights compared to non-DEGs). Color map compares particular types of response to Runx perturbation as compared to other responses to perturbation, among the DEGs with a given site type. Colors depict z-scores (standardized residuals), calculated for relative enrichment of a given association within the DEG groups. For example, dark cyan indicates that genes linked to a given site Group are especially positively enriched for the indicated response type. See Methods for how the non-DEGs and the core DEGs were defined. n of non-DEGs = 9471, n of OE only activated genes = 316, n of KO only activated genes 65, n of OE&KO activated genes = 100, n of OE only inhibited genes = 166, n of KO only inhibited genes 135, n of OE&KO inhibited genes = 46. *** Z score of Group 1 = -2.53, ** Z score of Group 1 = 2.22; *Z score of Group 3 = -3.36, * Z score of Group 3 = 1.94; *** Z score of Group 2b = 3.52; *Z score of Group 2a = -4.09, * Z score of Group 2a = 1.90. * |z-score|> 1.9599; ** |z-score| > 2.5758; *** |z-score| > 3.2905.b, Association of different groups of Runx binding sites with Runx target genes: among all genes in different categories that have any associated Runx sites, plots show what groups of Runx sites are most abundant. Violin plots show percent of each group of Runx peaks among total number of Runx peaks around a given gene. Runx peak groups are presented in an order of both empty vector-control and Runx1 OE binding sites (Group 1 and Group 3), preferentially occupied by Runx1 OE (Group 2b and Group 4), and post-commitment sites not occupied by experimental conditions in Phase 1 (Group 2a). Thin vertical black lines mark minima to maxima value range and thick vertical black boxes show 25th to 75th percentiles range. The white circles indicate median values. Group 3 sites are common in all DEG types but only Group 4cl are frequent with non-DEGs. c, Diagrams show a schematic summary of different groups of Runx peaks found commonly near Runx DEGs and Runx non-DEGs (Runx-independent). Note that each gene can possess multiple types of Runx peaks.

Source data

Extended Data Fig. 10 Runx factors function together with other TFs and control distinct sets of T-developmental genes.

a, Gene regulatory network analysis strategy using pySCENIC is shown. Cells were grouped by Runx perturbation condition and pseudotime category as shown to compute predicted target gene activity using SCENIC (pySCENIC, see Methods). b, SCENIC-predicted regulon activities for indicated TFs are represented as a heatmap. The expressed regulons scoring adj P < 1e-10 from more than two different pairwise comparisons using Kolmogorov–Smirnov tests were selected to display. c, The members of each regulon were overlapped with Runx DEGs defined by KO and/or OE from Fig. 3e. Then the numbers of overlapping predicted input regulons were enumerated per functionally responding Runx target gene or per non-DEG, and the results displayed as cumulative density functions. KS test P values were calculated by comparing Runx-activated or Runx-repressed DEGs with non-DEGs. Activated genes’ P = 1.55e-15, inhibited genes’ P = 8.88e-16. d, Curated Runx DEGs regulon memberships predicting input relationships are displayed as matrices. Colored cells in matrix indicate that a given Runx DEG (rows) is also a member of a given regulon (columns). Blue; Runx-activated genes, orange; Runx-inhibited genes. e-h, Area-proportional Venn diagrams display overlap patterns found between Runx DEGs with previously characterized functional targets of the indicated TFs. For Runx DEGs, genes activated (blue) or inhibited (orange) by Runx1 OE vs. Runx KO are each shown. Informative genes overlapping different classes of functionally responsive Runx DEGs are listed in different colored fonts: overlaps with Core-responsive DEGs showing reciprocal effects of Runx1 OE and KO (red); overlaps with DEGs defined by Runx1 OE-responses only (green); and overlaps with DEGs defined by Runx KO-responses only (blue) are listed. Comparisons between e, PU.1 target genes, f, GATA3 target genes, g, TCF1 target genes, and h, Bcl11b target genes are shown. For a simplified version, see Fig. 7c.

Source data

Supplementary information

Supplementary Information

Legends to Supplementary Tables 1–7, Notes 1–3 and References.

Reporting Summary

Supplementary Tables

Supplementary Tables 1–4 and 7. Supplementary Table 1. Runx-sensitive genes defined by scRNA-seq. Differentially expressed genes in control versus Runx1/Runx3 dKO or control versus Runx1 OE groups are shown. ‘Runx Core DEGs’ marks whether a given gene is sensitive to both Runx dKO and Runx1 OE. ‘Category’ indicates whether a given gene responded to Runx dKO and/or OE. ‘dn’ means downregulated in comparison to control cells and ‘up’ means upregulated in comparison to control cells. ‘Developmentally dynamic category’ is determined by differential gene expression analysis comparing cells in cluster 2 (early) versus cluster 1 (late). If a gene is significantly highly expressed in cluster 2, the gene is marked as a phase 1 DEG. If a gene is significantly upregulated in cluster 1, the gene is annotated as phase 2 DEG. Supplementary Table 2. scRNA-seq versus bulk RNA-seq comparison and Runx occupancy annotation. DEGs determined by scRNA-seq and previously reported bulk RNA-seq data were annotated. Previously reported bulk RNA-seq included two different timepoints of Runx1/Runx3 dKO. ‘Phase 1 bulk’ was measured by introducing gRunx1/Runx3 before T cell lineage commitment for 3 d (identical to the d3 post-infection timepoint in this study). ‘Phase 2 bulk’ was measured by deleting Runx1/Runx3 after 10 d of OP9-Dll1 co-culture (post-commitment) by introducing gRNA for 3 d. If a gene is scored as a DEG by an indicated method and timepoint, it was marked as ‘1’; if a gene was categorized as a non-DEG, it was marked as ‘0’. In addition, the numbers of annotated non-promoter Runx peaks in the indicated group (group 1, group 2a, group 2b, group 3, group 4a and group 4b) for each gene are marked. Supplementary Table 3. Total list of DEGs in ‘OE-diverging-UMAP2 window’ from Fig. 4. DEGs in control cells versus Runx1 OE cells within the UMAP2 values ranging from −30 to 5 are listed. Average log2fold change is calculated by comparing control/OE cells (genes expressed highly in control cells are positive). Supplementary Table 4. Differentially expressed genes in Spi1 hi cells from Fig. 4. DEGs in control cells versus Runx1 OE cells expressing Spi1 (count ≥3) are listed. Average log2fold change is calculated by comparing control/OE cells (genes expressed highly in control cells are positive). Supplementary Table 7. Summary of published datasets used in this work. Sample types, references and accession numbers are given.

Supplementary Table 5

SCENIC predicted Runx regulon members and their overlaps with Runx target genes. The putative target genes (regulon members) predicted by SCENIC in the scRNA-seq dataset are listed. The KS test P values for indicated comparisons are shown. The overlaps between Runx target genes and predicted regulon members are marked.

Supplementary Table 6

Comparison between Runx-sensitive genes and other TF-regulated genes presented in Fig. 7 and Extended Data Fig. 10. Runx DEGs defined by Runx1 OE and/or Runx1/Runx3 dKO were compared with the genes activated or inhibited by the indicated TFs that were previously reported.

Source data

Source Data Fig. 1

Statistical source data.

Source Data Fig. 2

Statistical source data.

Source Data Fig. 4

Statistical source data.

Source Data Fig. 5

Statistical source data.

Source Data Fig. 6

Statistical source data.

Source Data Fig. 7

Statistical source data.

Source Data Extended Data Fig. 1

Statistical source data.

Source Data Extended Data Fig. 2

Statistical source data.

Source Data Extended Data Fig. 3

Statistical source data.

Source Data Extended Data Fig. 4

Statistical source data.

Source Data Extended Data Fig. 6

Statistical source data.

Source Data Extended Data Fig. 7

Statistical source data.

Source Data Extended Data Fig. 9

Statistical source data.

Source Data Extended Data Fig. 10

Statistical source data.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shin, B., Zhou, W., Wang, J. et al. Runx factors launch T cell and innate lymphoid programs via direct and gene network-based mechanisms. Nat Immunol 24, 1458–1472 (2023). https://doi.org/10.1038/s41590-023-01585-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41590-023-01585-z

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing