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:

Reconstituting human somitogenesis in vitro

Nature volume 614, pages 509–520 (2023)Cite this article

Subjects

Abstract

The segmented body plan of vertebrates is established during somitogenesis, a well-studied process in model organisms; however, the details of this process in humans remain largely unknown owing to ethical and technical limitations. Despite recent advances with pluripotent stem cell-based approaches1,2,3,4,5, models that robustly recapitulate human somitogenesis in both space and time remain scarce. Here we introduce a pluripotent stem cell-derived mesoderm-based 3D model of human segmentation and somitogenesis—which we termed ‘axioloid’—that captures accurately the oscillatory dynamics of the segmentation clock and the morphological and molecular characteristics of sequential somite formation in vitro. Axioloids show proper rostrocaudal patterning of forming segments and robust anterior–posterior FGF–WNT signalling gradients and retinoic acid signalling components. We identify an unexpected critical role of retinoic acid signalling in the stabilization of forming segments, indicating distinct, but also synergistic effects of retinoic acid and extracellular matrix on the formation and epithelialization of somites. Comparative analysis demonstrates marked similarities of axioloids to the human embryo, further validated by the presence of a Hox code in axioloids. Finally, we demonstrate the utility of axioloids for studying the pathogenesis of human congenital spine diseases using induced pluripotent stem cells with mutations in HES7 and MESP2. Our results indicate that axioloids represent a promising platform for the study of axial development and disease in humans.

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: Generation of axioloids from human pluripotent stem cells.
Fig. 2: scRNA-seq characterization of human axioloids.
Fig. 3: Signalling pathways and the effect of Matrigel on axioloids.
Fig. 4: Effect of RA signalling on axioloid morphogenesis.
Fig. 5: Molecular comparison of axioloids with human embryos.
Fig. 6: Molecular and functional characterization of an axioloid model of SDV.

Similar content being viewed by others

Data availability

All scRNA-seq data and CUT&Tag data used for this study have been deposited in the NCBI Gene Expression Omnibus under accession number GSE199576.  Source data are provided with this paper.

Code availability

Computational codes and scripts used in this study are available at GitHub (https://github.com/Alev-Lab/axioloid-manuscript.git).

References

  1. Matsuda, M. et al. Recapitulating the human segmentation clock with pluripotent stem cells. Nature 580, 124–129 (2020).

    Article  CAS  PubMed  ADS  Google Scholar 

  2. van den Brink, S. C. et al. Single-cell and spatial transcriptomics reveal somitogenesis in gastruloids. Nature 582, 405–409 (2020).

    Article  PubMed  ADS  Google Scholar 

  3. Veenvliet, J. V. et al. Mouse embryonic stem cells self-organize into trunk-like structures with neural tube and somites. Science 370, eaba4937 (2020).

    Article  CAS  PubMed  Google Scholar 

  4. Diaz-Cuadros, M. et al. In vitro characterization of the human segmentation clock. Nature 580, 113–118 (2020).

    Article  CAS  PubMed  PubMed Central  ADS  Google Scholar 

  5. Moris, N. et al. An in vitro model of early anteroposterior organization during human development. Nature 582, 410–415 (2020).

    Article  CAS  PubMed  ADS  Google Scholar 

  6. Chu, L. F. et al. An in vitro human segmentation clock model derived from embryonic stem cells. Cell Rep. 28, 2247–2255.e2245 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Wilson, V., Rashbass, P. & Beddington, R. S. Chimeric analysis of T (Brachyury) gene function. Development 117, 1321–1331 (1993).

    Article  CAS  PubMed  Google Scholar 

  8. McGrew, M. J., Dale, J. K., Fraboulet, S. & Pourquie, O. The lunatic Fringe gene is a target of the molecular clock linked to somite segmentation in avian embryos. Curr. Biol. 8, 979–982 (1998).

    Article  CAS  PubMed  Google Scholar 

  9. Prince, V. E. et al. Zebrafish lunatic fringe demarcates segmental boundaries. Mech. Dev. 105, 175–180 (2001).

    Article  CAS  PubMed  Google Scholar 

  10. Evrard, Y. A., Lun, Y., Aulehla, A., Gan, L. & Johnson, R. L. lunatic fringe is an essential mediator of somite segmentation and patterning. Nature 394, 377–381 (1998).

    Article  CAS  PubMed  ADS  Google Scholar 

  11. Neidhardt, L. M., Kispert, A. & Herrmann, B. G. A mouse gene of the paired-related homeobox class expressed in the caudal somite compartment and in the developing vertebral column, kidney and nervous system. Dev. Genes Evol. 207, 330–339 (1997).

    Article  CAS  PubMed  Google Scholar 

  12. Kraus, F., Haenig, B. & Kispert, A. Cloning and expression analysis of the mouse T-box gene Tbx18. Mech. Dev. 100, 83–86 (2001).

    Article  CAS  PubMed  Google Scholar 

  13. Pourquie, O. The segmentation clock: converting embryonic time into spatial pattern. Science 301, 328–330 (2003).

    Article  CAS  PubMed  ADS  Google Scholar 

  14. Holley, S. A., Geisler, R. & Nusslein-Volhard, C. Control of her1 expression during zebrafish somitogenesis by a delta-dependent oscillator and an independent wave-front activity. Genes Dev 14, 1678–1690 (2000).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Saga, Y. & Takeda, H. The making of the somite: molecular events in vertebrate segmentation. Nat. Rev. Genet. 2, 835–845 (2001).

    Article  CAS  PubMed  Google Scholar 

  16. Bessho, Y. et al. Dynamic expression and essential functions of Hes7 in somite segmentation. Genes Dev. 15, 2642–2647 (2001).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Palmeirim, I., Henrique, D., Ish-Horowicz, D. & Pourquie, O. Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesis. Cell 91, 639–648 (1997).

    Article  CAS  PubMed  Google Scholar 

  18. Ishimatsu, K., Takamatsu, A. & Takeda, H. Emergence of traveling waves in the zebrafish segmentation clock. Development 137, 1595–1599 (2010).

    Article  CAS  PubMed  Google Scholar 

  19. Cooke, J. & Zeeman, E. C. A clock and wavefront model for control of the number of repeated structures during animal morphogenesis. J. Theor. Biol. 58, 455–476 (1976).

    Article  CAS  PubMed  ADS  Google Scholar 

  20. Kondrashov, N. et al. Ribosome-mediated specificity in Hox mRNA translation and vertebrate tissue patterning. Cell 145, 383–397 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Olivera-Martinez, I., Harada, H., Halley, P. A. & Storey, K. G. Loss of FGF-dependent mesoderm identity and rise of endogenous retinoid signalling determine cessation of body axis elongation. PLoS Biol. 10, e1001415 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Diez del Corral, R. et al. Opposing FGF and retinoid pathways control ventral neural pattern, neuronal differentiation, and segmentation during body axis extension. Neuron 40, 65–79 (2003).

    Article  CAS  PubMed  Google Scholar 

  23. Olivera-Martinez, I. & Storey, K. G. Wnt signals provide a timing mechanism for the FGF-retinoid differentiation switch during vertebrate body axis extension. Development 134, 2125–2135 (2007).

    Article  CAS  PubMed  Google Scholar 

  24. Moreno, T. A. & Kintner, C. Regulation of segmental patterning by retinoic acid signaling during Xenopus somitogenesis. Dev. Cell 6, 205–218 (2004).

    Article  CAS  PubMed  Google Scholar 

  25. Dubrulle, J., McGrew, M. J. & Pourquie, O. FGF signaling controls somite boundary position and regulates segmentation clock control of spatiotemporal Hox gene activation. Cell 106, 219–232 (2001).

    Article  CAS  PubMed  Google Scholar 

  26. Aulehla, A. et al. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev. Cell 4, 395–406 (2003).

    Article  CAS  PubMed  Google Scholar 

  27. Aulehla, A. et al. A beta-catenin gradient links the clock and wavefront systems in mouse embryo segmentation. Nat. Cell Biol. 10, 186–193 (2008).

    Article  CAS  PubMed  Google Scholar 

  28. Gyllborg, D. et al. Hybridization-based in situ sequencing (HybISS) for spatially resolved transcriptomics in human and mouse brain tissue. Nucleic Acids Res. 48, e112 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Aulehla, A. & Pourquie, O. Signaling gradients during paraxial mesoderm development. Cold Spring Harb. Perspect. Biol. 2, a000869 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  30. Vermot, J. et al. Retinoic acid controls the bilateral symmetry of somite formation in the mouse embryo. Science 308, 563–566 (2005).

    Article  CAS  PubMed  ADS  Google Scholar 

  31. Sato, Y. et al. Notch mediates the segmental specification of angioblasts in somites and their directed migration toward the dorsal aorta in avian embryos. Dev. Cell 14, 890–901 (2008).

    Article  CAS  PubMed  Google Scholar 

  32. Jacob, M., Christ, B., Jacob, H. J. & Poelmann, R. E. The role of fibronectin and laminin in development and migration of the avian Wolffian duct with reference to somitogenesis. Anat. Embryol. 183, 385–395 (1991).

    Article  CAS  Google Scholar 

  33. Mills, C. L., Ariyo, O., Yamada, K. M., Lash, J. W. & Bellairs, R. Evidence for the involvement of receptors for fibronectin in the promotion of chick tail segmentation. Anat. Embryol. 182, 425–434 (1990).

    Article  CAS  Google Scholar 

  34. Page, M. Changing patterns of cytokeratins and vimentin in the early chick embryo. Development 105, 97–107 (1989).

    Article  CAS  PubMed  Google Scholar 

  35. Burgess, R., Rawls, A., Brown, D., Bradley, A. & Olson, E. N. Requirement of the paraxis gene for somite formation and musculoskeletal patterning. Nature 384, 570–573 (1996).

    Article  CAS  PubMed  ADS  Google Scholar 

  36. Wiggan, O., Fadel, M. P. & Hamel, P. A. Pax3 induces cell aggregation and regulates phenotypic mesenchymal-epithelial interconversion. J. Cell Sci. 115, 517–529 (2002).

    Article  CAS  PubMed  Google Scholar 

  37. Rowton, M. et al. Regulation of mesenchymal-to-epithelial transition by PARAXIS during somitogenesis. Dev. Dyn. 242, 1332–1344 (2013).

    Article  CAS  PubMed  Google Scholar 

  38. Vermot, J. & Pourquie, O. Retinoic acid coordinates somitogenesis and left-right patterning in vertebrate embryos. Nature 435, 215–220 (2005).

    Article  CAS  PubMed  ADS  Google Scholar 

  39. Xu, Y. et al. A single-cell transcriptome atlas of human early embryogenesis. Preprint at bioRxiv https://doi.org/10.1101/2021.11.30.470583 (2022).

  40. Stuart, T. et al. Comprehensive integration of single-cell data. Cell 177, 1888–1902.e1821 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Nguyen, P. D. et al. Haematopoietic stem cell induction by somite-derived endothelial cells controlled by meox1. Nature 512, 314–318 (2014).

    Article  CAS  PubMed  ADS  Google Scholar 

  42. Pardanaud, L. & Dieterlen-Lievre, F. Manipulation of the angiopoietic/hemangiopoietic commitment in the avian embryo. Development 126, 617–627 (1999).

    Article  CAS  PubMed  Google Scholar 

  43. Lippmann, E. S. et al. Deterministic HOX patterning in human pluripotent stem cell-derived neuroectoderm. Stem Cell Rep. 4, 632–644 (2015).

    Article  CAS  Google Scholar 

  44. Beccari, L. et al. Multi-axial self-organization properties of mouse embryonic stem cells into gastruloids. Nature 562, 272–276 (2018).

    Article  CAS  PubMed  ADS  Google Scholar 

  45. Matsumiya, M., Tomita, T., Yoshioka-Kobayashi, K., Isomura, A. & Kageyama, R. ES cell-derived presomitic mesoderm-like tissues for analysis of synchronized oscillations in the segmentation clock. Development 145, dev156836 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  46. Wahl, M. B., Deng, C., Lewandoski, M. & Pourquie, O. FGF signaling acts upstream of the NOTCH and WNT signaling pathways to control segmentation clock oscillations in mouse somitogenesis. Development 134, 4033–4041 (2007).

    Article  CAS  PubMed  Google Scholar 

  47. Dubrulle, J. & Pourquie, O. fgf8 mRNA decay establishes a gradient that couples axial elongation to patterning in the vertebrate embryo. Nature 427, 419–422 (2004).

    Article  CAS  PubMed  ADS  Google Scholar 

  48. Boulet, A. M. & Capecchi, M. R. Signaling by FGF4 and FGF8 is required for axial elongation of the mouse embryo. Dev. Biol. 371, 235–245 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Sparrow, D. B., Guillen-Navarro, E., Fatkin, D. & Dunwoodie, S. L. Mutation of HAIRY-AND-ENHANCER-OF-SPLIT-7 in humans causes spondylocostal dysostosis. Hum. Mol. Genet. 17, 3761–3766 (2008).

    Article  CAS  PubMed  Google Scholar 

  50. Whittock, N. V. et al. Mutated MESP2 causes spondylocostal dysostosis in humans. Am. J. Hum. Genet. 74, 1249–1254 (2004).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Takahashi, Y. et al. Mesp2 initiates somite segmentation through the Notch signalling pathway. Nat. Genet. 25, 390–396 (2000).

    Article  CAS  PubMed  Google Scholar 

  52. Takahashi, Y. et al. Transcription factors Mesp2 and Paraxis have critical roles in axial musculoskeletal formation. Dev. Dyn. 236, 1484–1494 (2007).

    Article  CAS  PubMed  Google Scholar 

  53. Sanaki-Matsumiya, M. et al. Periodic formation of epithelial somites from human pluripotent stem cells. Nat. Commun. 13, 2325 (2022).

    Article  CAS  PubMed  PubMed Central  ADS  Google Scholar 

  54. Budjan, C. et al. Paraxial mesoderm organoids model development of human somites. eLife 11, e68925 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Cooper, F. et al. Rostrocaudal patterning and neural crest differentiation of human pre-neural spinal cord progenitors in vitro. Stem Cell Rep. 17, 894–910 (2022).

    Article  CAS  Google Scholar 

  56. Grosswendt, S. et al. Epigenetic regulator function through mouse gastrulation. Nature 584, 102–108 (2020).

    Article  CAS  PubMed  PubMed Central  ADS  Google Scholar 

  57. Okita, K. et al. A more efficient method to generate integration-free human iPS cells. Nat. Methods 8, 409–412 (2011).

    Article  CAS  PubMed  Google Scholar 

  58. Takahashi, K. et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861–872 (2007).

    Article  CAS  PubMed  Google Scholar 

  59. Yamanaka, Y., Hamidi, S., Ericsson, R., Munira, S. & Alev, C. Derivation of axioloids, an in vitro model of somitogenesis and axial development from human pluripotent stem cells. Protoc. Exch. https://doi.org/10.21203/rs.3.pex-2114/v1 (2023).

  60. Cork, R. J. & Gasser, R. F. The Virtual Human Embryo Project: a resource for the study of human embryology. FASEB J. 26, 14.14 (2012).

    Article  Google Scholar 

  61. Sharpe, J. et al. Optical projection tomography as a tool for 3D microscopy and gene expression studies. Science 296, 541–545 (2002).

    Article  CAS  PubMed  ADS  Google Scholar 

  62. Kerwin, J. et al. 3 dimensional modelling of early human brain development using optical projection tomography. BMC Neurosci. 5, 27 (2004).

    Article  PubMed  PubMed Central  Google Scholar 

  63. Kerwin, J. et al. The HUDSEN Atlas: a three-dimensional (3D) spatial framework for studying gene expression in the developing human brain. J. Anat. 217, 289–299 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. Gering, D. T. et al. An integrated visualization system for surgical planning and guidance using image fusion and an open MR. J. Magn. Reson. Imaging 13, 967–975 (2001).

    Article  CAS  PubMed  Google Scholar 

  65. Hama, H. et al. ScaleS: an optical clearing palette for biological imaging. Nat. Neurosci. 18, 1518–1529 (2015).

    Article  CAS  PubMed  Google Scholar 

  66. Choi, H. M. T. et al. Third-generation in situ hybridization chain reaction: multiplexed, quantitative, sensitive, versatile, robust. Development 145, dev165753 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  67. Masamizu, Y. et al. Real-time imaging of the somite segmentation clock: revelation of unstable oscillators in the individual presomitic mesoderm cells. Proc. Natl Acad. Sci. USA 103, 1313–1318 (2006).

    Article  CAS  PubMed  PubMed Central  ADS  Google Scholar 

  68. Isomura, A., Ogushi, F., Kori, H. & Kageyama, R. Optogenetic perturbation and bioluminescence imaging to analyze cell-to-cell transfer of oscillatory information. Genes Dev. 31, 524–535 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).

    Article  CAS  PubMed  Google Scholar 

  70. Kaya-Okur, H. S. et al. CUT&Tag for efficient epigenomic profiling of small samples and single cells. Nat. Commun. 10, 1930 (2019).

    Article  PubMed  PubMed Central  ADS  Google Scholar 

  71. Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17, 10–12 (2011).

    Article  Google Scholar 

  72. 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 

  73. Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  74. Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

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

    Article  PubMed  PubMed Central  Google Scholar 

  76. Wolock, S. L., Lopez, R. & Klein, A. M. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst. 8, 281–291.e289 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  77. Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587.e3529 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  78. Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 411–420 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  79. Hafemeister, C. & Satija, R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. Genome Biol. 20, 296 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  80. Bray, N. L., Pimentel, H., Melsted, P. & Pachter, L. Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34, 525–527 (2016).

    Article  CAS  PubMed  Google Scholar 

  81. Bergen, V., Lange, M., Peidli, S., Wolf, F. A. & Theis, F. J. Generalizing RNA velocity to transient cell states through dynamical modeling. Nat. Biotechnol. 38, 1408–1414 (2020).

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank G. Sheng, K.T. Kim and S. Goulas for critical reading and constructive suggestions for the manuscript; members of the single cell core facility (SignAC) at ASHBi and S. Terakura for help with scRNA-seq experiments; K. Terai and other members of the Matsuda laboratory at Kyoto University for help with bioluminescence live-imaging experiments; members of the iCeMS Analysis Center and Y. Tomoda and S. Kihara from the CiRA imaging core facility for support with microscopy and high content live imaging; Y. Imoto from ASHBi for help with computational analysis and graph generation; Y. Arai and members of the ASHBi office for administrative support. This work was supported by Naito Foundation Scientific Research Grant to C.A.; Takeda Science Foundation Grant to C.A.; Japan Agency for Medical Research and Development (AMED) grant number JP20bm0704050 to C.A.; the ISHIZUE 2020 of Kyoto University Research Development Program to C.A.; JST FOREST Program (Grant Number JPMJFR206C) to T.Y.; JST, CREST (grant number JPMJCR2023) to T.Y.; AMED-CREST (grant numbers JP22gm1310002 and JP22gm1310011) to T.Y.; the ASHBi Fusion Research Program to K.Y.-K., Y.Y., K.S., T.Y. and C.A. The human embryonic and fetal material was provided by the Joint MRC / Wellcome (MR/R006237/1) Human Developmental Biology Resource (www.hdbr.org). N.M is supported by the Francis Crick Institute, which receives its core funding from Cancer Research UK (FC011181), the UK Medical Research Council (FC011181) and the Wellcome Trust (FC011181), and a Royal Society research grant (RGS\R2\212082). A.M.-A., N.M. and C.A. were additionally supported by the UK Medical Research Council and Japanese Agency for Medical Research and Development AMED (MR/V005367/2). ASHBi is supported by the World Premier International Research Center Initiative (WPI), MEXT, Japan.

Author information

Author notes

  1. These authors contributed equally: Yoshihiro Yamanaka, Sofiane Hamidi

Authors and Affiliations

  1. Institute for the Advanced Study of Human Biology (ASHBi), Kyoto University, Kyoto, Japan

    Yoshihiro Yamanaka, Sofiane Hamidi, Kumiko Yoshioka-Kobayashi, Sirajam Munira, Kazunori Sunadome, Yi Zhang, Yuzuru Kurokawa, Rolf Ericsson, Ai Mieda, Takuya Yamamoto, Taro Tsujimura & Cantas Alev

  2. Department of Life Science Frontiers, Center for iPS Cell Research and Application (CiRA), Kyoto University, Kyoto, Japan

    Kazunori Sunadome & Takuya Yamamoto

  3. The Francis Crick Institute, London, UK

    Jamie L. Thompson & Naomi Moris

  4. HDBR, Newcastle University, Newcastle Upon Tyne, UK

    Janet Kerwin & Steven Lisgo

  5. Medical-Risk Avoidance Based on iPS Cells Team, RIKEN Center for Advanced Intelligence Project (AIP), Kyoto, Japan

    Takuya Yamamoto

  6. Department of Medicine and Life Sciences, Universitat Pompeu Fabra and ICREA, Barcelona, Spain

    Alfonso Martinez-Arias

Authors
  1. Yoshihiro Yamanaka
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sofiane Hamidi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kumiko Yoshioka-Kobayashi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sirajam Munira
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Kazunori Sunadome
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yi Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yuzuru Kurokawa
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Rolf Ericsson
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Ai Mieda
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Jamie L. Thompson
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Janet Kerwin
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Steven Lisgo
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Takuya Yamamoto
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Naomi Moris
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Alfonso Martinez-Arias
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Taro Tsujimura
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Cantas Alev
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Y.Y., K.Y.-K., S.M., S.H. and C.A. developed the human axioloid induction protocol and performed molecular and functional analysis of samples. A.M.-A. and N.M. provided critical scientific feedback and frequent discussions. Y.Z. helped with molecular characterization of samples and data analysis. K.S. and T.Y. performed HybISS analysis of samples with the help of Y.Y. and S.H. R.E. helped with data analysis, manuscript preparation and figure assembly. J.L.T. repeated axioloid experiments in the UK, with support from N.M. J.K. and S.L. provided data for CS10 and CS11 human embryo samples. S.H. quantified somite dimensions in axioloids and human embryos. A.M., Y.K. and T.T. performed and analysed CUT&Tag experiments. Y.Y. performed and T.T., Y.K. and T.Y. analysed scRNA-seq data. C.A. conceived, designed and supervised the study. C.A. analysed and interpreted the data and wrote the manuscript, incorporating feedback from the co-authors. All authors discussed and commented on the manuscript and agreed on the presented results.

Corresponding author

Correspondence to Cantas Alev.

Ethics declarations

Competing interests

Y.Y., S.H., S.M. and C.A., together with Kyoto University, have filed a provisional patent application (US63/326,611) covering the generation and use of axioloids. The other authors declare no competing interests.

Peer review

Peer review information

Nature thanks Jesse Veenvliet and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

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

Extended data figures and tables

Extended Data Fig. 1 Morphological and molecular characterization of human axioloids.

a–c, Bright field images of a, elongating axioloids at 24 h, 48 h, and 72 h, followed by images of axioloids after MG embedding at 96 h and 120 h or b,c, without MG embedding in axioloids derived from 409B2 or 201B7 Luc respectively. d, Representative bright field images reflecting the different morphology-based categories (Cat#1: straight, minimal curvature, clear segments; Cat#2: curved but clear segments; Cat#3: very curved, segment borders are still distinguishable; Cat#4: not properly elongated or completely collapsed; segment borders not clear) of 409B2-derived MG embedded axioloids at 96 h and 120 h (3 independent experiments, n = 128 axioloids, mean ±SD). e, Serial images of a forming axioloid at 5 h intervals, from 74 h to 119 h (extracted from Supplementary Video 2). Yellow arrowheads pinpoint to areas where segmentation is ongoing, whereas green arrowheads highlight the areas where segmentation is completed. f, Periodicity of segmentation based on live-cell imaging observations for 409B2 (corresponding serial images shown in Fig. 1c, 3 independent experiments, n = 9, mean ±SD) and 201B7 (corresponding serial images shown in e, 3 independent experiments, n = 9, mean ±SD). g, Axioloid length (antero-posterior) at 24 h, 48 h, 72 h, and at 96 h and 120 h with or without MG embedding (3 independent experiments; no MG 24 h, 48 h and 72 h n = 18, and -MG 96 h n = 14, +MG 96 h n = 18, and -MG 120 h n = 13, +MG 120 h n = 18). h–r, Immunofluorescence staining and quantification of axioloids. h,j,n,o,q, Representative immunofluorescence images of F-actin (Phalloidin) in gray, TBXT in cyan, and MEOX1 in red; stained axioloids at n, 72 h no MG, h,j, 96 h and 120 h with MG embedding and o,q, 96 h and 120 h without MG embedding. i,k,p, and r, Corresponding quantification along the posterior to anterior axis of TBXT (cyan line) and MEOX1 (in red) signal intensity at i, 96 h and 120 h (3 independent experiments, n = 9), k, 96 h (3 independent experiments, n = 14) and 120 h (4 independent experiments, n = 16) with MG embedding, and p,r, 96 h (409B2 3 independent experiments, n = 9 and 201B7 3 independent experiments, n = 15) and 120 h (409B2 3 independent experiments, n = 9 and 201B7 3 independent experiments, n = 11) without MG embedding. l–m, Segment length was measured based on Phalloidin staining done in h and j in both cell lines 409B2 (3 independent experiments, n = 9) and 201B7 Luc (3 independent experiments, n = 9). Single channel images shown in h at 96 h corresponds to the merged channel image shown in Fig. 1e. Boxes in g,l and m represent the first and last 4th percentile intervals, whiskers show the minimum and maximum values within that range, middle lines the median, and dots represent individual data points. Lines in i,k,p and r correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Scale bar is 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 2 Assessment of apicobasal polarity, developmental protein- & gene expression patterns, rostrocaudal patterning, traveling wave front of HES7 oscillatory activity & segmentation in human axioloids embedded in MG.

a–d, Immunofluorescence staining of MG embedded axioloids. a, and b, High magnification images (63X) of a single segment at 120 h with F-actin (Phalloidin) in gray, aPKC and FN1 in cyan, and MEOX1 in red in a, 409B2 and b, 201B7 Luc iPS cell line-derived axioloids. c, Images of F-actin (Phalloidin) in yellow, TBXT in cyan and SOX2 in magenta at 25X magnification and d, magnified view (63X) of the stained area in c. e, Images of F-actin (Phalloidin) in gray, TBX6 in cyan and SOX2 in red. f–q, HCR staining and corresponding signal quantification of stained axioloids; f, and g, HCR staining of MSGN1 in cyan, TCF15 in magenta and RIPPLY2 in yellow; h, and i, HCR staining of RIPPLY2 in yellow, LFNG in cyan and HES7 in magenta; red arrowheads highlight the stripe-like staining pattern of LFNG in the posterior half of each somite. j–q, HCR staining of MESP2 in yellow, UNCX in cyan and TBX18 in magenta, and corresponding signal intensity measurements along the posterior to anterior axis normalized to the position of the MESP2 signal peak of axioloids embedded in MG at 96 h and 120 h respectively in 2 different cell lines shown in j,k, 409B2 (4 independent experiments, n = 10 and 3 independent experiments, n = 9) and l,m, 201B7 Luc (3 independent experiments, n = 9 and 3 independent experiments, n = 10) and axioloids without MG embedding at 96 h in 2 cell lines shown in n,o, 409B2 (3 independent experiments, n = 12) and p,q, 201B7 Luc (3 independent experiments, n = 11). Single channel images shown in e,f,h, and j top panel, correspond to the merged channel images shown in Fig. 1f-i. r, and s, Annotated serial images of a forming axioloid with HES7:Luciferase signal overlayed in green (extracted from Supplementary Video 3). Colored dotted lines mark the furthermost anterior position reached by each HES7 oscillation wave of gene expression; identical colored arrowheads mark this position overtime, yellow first oscillation, red second, blue third, orange forth. s, Image at 24 h shows that each HES7:Luciferase expression wavefront position corresponds to the area of a formed segment. t, Average HES7:Luciferase intensity measurements overtime (4 independent experiments, n = 8) and u, corresponding signal periodicity measurement (4 independent experiments, n = 8, mean ±SD). Images in a,b,d,f and h are representative of 2 independent experiments; in c, and e, of 3 independent experiments. Lines in g,I,k,m,o,q and t correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Scale bars in a,b and d 50 µm, others are 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 3 Single cell RNA-seq analysis of human axioloids.

a, UMAP projection of scRNA-seq datasets of axioloids at 48 h, 72 h, 96 h and 120 h, colored by inferred cell cycle phases (G1, G2M, S). b, G2M.Score and S.Score of the cells in each cluster of Fig. 2b. c, Proportions of cell types in axioloids with and without MG for both 96 h and 120 h timepoints. d, Averaged expression levels of ribosomal protein genes in each cluster of Fig. 2b. e, Transition of TB marker gene expression along the time course. f, Expression levels of TBXT and SOX2 in each cell are plotted for the three TB clusters (E-TB, M-TB and L-TB, standing for early, mid and late tailbud respectively). g, Neuromesodermal progenitor (NMP) (left) and neuroectoderm (NE) (right) module scores of reported marker gene expression3 in analyzed single cells. h, Time course expression changes of previously reported NMP and neural marker genes55 in three TB clusters. i, Selected genes differentially expressed in early (E-TB) compared to late (L-TB) tailbud cell populations; top genes are highly expressed in the L-TB, bottom genes show higher expression in the E-TB. j, and k, UMAP plots of human axioloids at 72 h in j and 120 h (with Matrigel) in k colored by identified clusters. Arrows show RNA velocity.

Source Data

Extended Data Fig. 4 FGF and WNT signaling pathway expression and activity in human MG embedded axioloids.

a–b, Expression patterns of FGF and WNT signaling pathway associated transcripts in human MG exposed axioloids at 96 h of culture (24 h after embedding into MG) along the pseudotime axis. Effectors and negative regulators of both pathways are included. c–g, HCR staining images and signal quantification of MG embedded axioloids at 96 h and 120 h derived from c, 409B2 and d–g, 201B7 Luc iPSC lines. Shown images are representative of 3 independent experiments. c–e, HCR staining of MESP2 in yellow, FGF8 in cyan and WNT3A in magenta and corresponding quantification along the posterior to anterior axis of the signal intensity normalized to the position of the WNT3A signal peak for d, and e, (96 h: 3 independent experiments, n = 8 and 120 h: 3 independent experiments, n = 7). f, HCR staining of LEF1 in cyan, AXIN2 in magenta and WNT3A in yellow and corresponding quantification (3 independent experiments, n = 9). g, HCR staining of DUSP6 in cyan, SPRY4 in magenta, and FGF8 in yellow and corresponding quantification (3 independent experiments, n = 9). Single channel images shown in c correspond to the merged channel images shown in Fig. 3a. h–o, HybISS based visualization and quantification of the spatial expression of WNT and FGF signaling pathway members along the posterior to anterior axis in human axioloids at 96 h (top n = 3) and 120 h (bottom n = 2) of culture; WNT signaling members shown in h–k and FGF signaling members shown in l–o; corresponding quantification of spatial expression patterns for WNT pathway related transcripts shown in i, and k, and for FGF pathway related transcripts shown in m, and o. p, Immunofluorescence staining of phospho-ERK (pERK) in cyan, beta-Catenin (CTNNB1) in magenta and TBX6 in yellow in 201B7 Luc-derived MG embedded axioloids at 96 h. q, Corresponding quantification of the signal intensity along the posterior to anterior axis (3 independent experiments, n = 8). r, Magnification of the posterior-most region of the axioloid shown in p. Lines in d,e,f,g and q correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Lines in i,k,m and o correspond to mean values; error bands represent the SD for each data point. Scale bar is 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 5 Expression gradients of RA signaling pathway members in human axioloids embedded in MG.

a, Pseudotime representation of expression of RA signaling pathway associated transcripts in human MG exposed axioloids at 96 h of culture (24 h after embedding into MG); effectors and negative regulators of the RA pathway gene expression patterns are arranged along pseudotime rank. b–f, HCR staining images and signal quantification of MG embedded axioloids at 96 h and 120 h derived from b, 409B2 and c–f, 201B7 Luc iPSC lines. Shown images are representative of at least two independent experiments. b–d, HCR staining of CYP26A1 in cyan, ALDH1A2 in magenta, RIPPLY2 in yellow of axioloids derived from b, 409B2 at 96 h and 120 h and 201B7 at c, 96 h and d, 120 h and related quantification along the posterior to anterior axis of the signal intensity. (96 h: 3 independent experiments, n = 6 and 120 h: 3 independent experiments, n = 7). e, HCR staining of +MG axioloid at 96 h of culture for RARA in cyan, CRABP2 in magenta and RIPPLY2 in yellow and corresponding quantification (2 independent experiments, n = 6) and f, HCR staining of +MG axioloid at 96 h of culture for RARG in cyan, RDH10 in magenta, and RIPPLY2 in yellow (2 independent experiments, n = 6). Single channel images shown in b correspond to the merged channel images shown in Fig. 3c. g, i, k, HybISS based visualization of the spatial distribution along the posterior to anterior axis of RA signaling pathway members in human axioloids at 96 h (top n = 3) and 120 h (bottom n = 2) of culture in 409B2 derived axioloids embedded in MG and h,j,l, corresponding quantification of spatial expression. Lines in c–f correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Lines in h,j and l correspond to mean values; error bands represent the SD for each data point. Scale bar is 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 6 Single cell RNA-seq analysis: identification of DEGs associated with the exposure of axioloids to MG.

a, UMAP projection of the integrated two replicates of axioloids at 96 h with MG, colored by the clusters of Fig. 2g. b, UMAP projection of the integrated two replicates of axioloids at 96 h with MG, colored by the clusters of Fig. 2b. Note that b, includes only replicate 1. c, Averaged expression levels of identified EC-like marker genes in each cluster of Fig. 2b. d, Expression levels of indicated genes of the same UMAP plot in a. e, and f, Differentially expressed genes between axioloids with and without MG at 96 h in PSM and TB, respectively, and f, a volcano plot of TB. Red and blue dots indicate up- and down-regulated genes by MG in both replicates respectively; expression changes were calculated using the FindMarkers function with a Wilcoxon rank sum test (two-sided). P-values adjusted using Bonferroni correction were plotted.

Source Data

Extended Data Fig. 7 Assessing the morphogenetic effects of retinoid signaling on human axioloids.

a, and b, Bright field images of axioloids at 96 h and 120 h after embedding in MG (Matrigel) only, +MG +RAL (retinal), +MG +ROL (retinol) or +MG +RA (retinoic acid) for a, 409B2 (3 independent experiments) and b, 201B7 Luc (3 independent experiments). c, Serial images of an elongating axioloid at 5 h intervals, from 74 h to 119 h (extracted from Supplementary Video 4). Yellow arrowheads pinpointing to areas where somite formation and segmentation is ongoing whereas red for 409B2 or green for 201B7 Luc arrowheads highlight the areas where segmentation is completed. d, and e, Immunofluorescence high magnification images (X63) of a single somite of axioloids embedded in +MG +RAL at 120 h with F-actin (Phalloidin) in gray, and from top to bottom aPKC, FN1 in cyan, and MEOX1 in red in d, 409B2 and e, 201B7 Luc iPSC lines; images are representative of two independent experiments. f, Length measurement of axioloids at 96 h and 120 h after embedding in MG only, +MG +ROL, +MG +RAL or +MG +RA for 201B7 Luc (3 independent experiments,  n = 18). g, Periodicity of somite segmentation based on live-cell imaging observations (4 independent experiments, n = 9, mean ±SD). h, Total number of somites in 201B7 derived axioloids embedded in +MG +ROL or +MG +RAL or +MG +RA at 120 h (3 independent experiments, n = 30 for ROL, n = 60 for RAL and RA). i–n, Representative images of immunofluorescence staining of F-actin (Phalloidin) in gray, TBXT in cyan, and MEOX1 in red and corresponding quantification of signal intensity of axioloids embedded in +MG +ROL shown in i–j, 201B7 Luc at 96 h (3 independent experiments, n = 7) and 120 h (3 independent experiments, n = 9). Immunofluorescence data of axioloids embedded in +MG +RAL shown in k–l, 201B7 Luc at 96 h (4 independent experiments, n = 13) and 120 h (4 independent experiments, n = 13). Immunofluorescence data of axioloids embedded in +MG +RA shown m–n, 201B7 Luc at 96 h (3 independent experiments, n = 15) and 120 h (4 independent experiments, n = 16). o–t, Representative images of HCR staining of MESP2 in yellow, UNCX in cyan and TBX18 in magenta, and corresponding signal intensity measurements along the posterior to anterior axis normalized to the position of the MESP2 signal peak of axioloids embedded in +MG +ROL shown in o–p, 201B7 at 96 h (3 independent experiments, n = 8) and at 120 h (3 independent experiments, n = 8); and in +MG +RAL shown in q–r, 201B7 Luc at 96 h (3 independent experiments, n = 9) and 120 h (3 independent experiments, n = 9). In situ hybridization data of axioloids embedded in +MG +RA shown in s–t, 201B7 Luc at 96 h (2 independent experiments, n = 5) and 120 h (2 independent experiments, n = 5). Lines in j,l,n,p,r and t correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Boxes in f and h represent the first and last 4th percentile intervals, whiskers show the minimum and maximum values within that range, middle line shows the median, and dots represent individual data points. Scale bar in d and e is 50 µm, others are 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 8 Molecular characterization of human axioloids exposed to agonists or inhibitors of retinoic acid (RA) signaling.

a, Brightfield images of 201B7 Luc-derived axioloids at 96 h (top) and 120 h (bottom) embedded in MG alone and supplemented with (from left to right) DMSO (2 independent experiments, n = 12 for 96 h and n = 13 for 120 h), BMS493 (a pan-RAR inverse agonist) (3 independent experiments, n = 18 for 96 h and n = 17 for 120 h), or +MG +RAL supplemented with DMSO (3 independent experiments, n = 9 for both 96 h & 120 h), BMS493 (3 independent experiments, n = 9 for both 96 h & 120 h), AGN193109 (a pan-RAR inhibitor) (3 independent experiments, n = 11 for somite number and n = 18 for length measurements for 96 h & 120 h) or ER50891 (a RARα-specific inhibitor) (3 independent experiments, n = 12 for somite number and n = 18 for length measurements for 96 h & 120 h). b, Corresponding axioloid length measurements and somite numbers for axioloids shown in a (axioloids embedded in MG only are not plotted). Boxes represent the first and last 4th percentile intervals, whiskers show the minimum and maximum values within that range, middle line the median, and dots represent individual data points. Somite numbers for +MG +RAL (3 independent experiments, n = 9 for both 96 h & 120 h) are shown in upper panel. Length measurements for +MG +BMS (3 independent experiments, n = 18 for 96 h and n = 17 for 120 h) and +MG + DMSO (3 independent experiments, n = 9 for 96 h and n = 9 for 120 h) are shown in the lower panel. c, Serial images of forming +MG +RAL axioloids at 5 h intervals, from 74 h to 119 h in the absence or presence of RA signaling inhibitors (extracted from Supplementary Video 6). Red arrowheads pinpointing to areas where segmentation is ongoing whereas blue arrowheads highlight the areas where segmentation is completed. d–m, Immunofluorescence and in situ hybridization-based characterization of axioloids treated with different RA signaling inhibitors. d,j, and l, Immunofluorescence staining of axioloids embedded in +MG +RAL supplemented with d, BMS493, j, AGN193109 and l, ER50891 at 120 h with F-actin (Phalloidin) in white, TBXT in cyan, FN1 in yellow and MEOX1 in red. e, Signal intensity quantification of d, along the posterior to anterior axis (3 independent experiments, n = 11). f,h,k, and m, HCR staining of UNCX in cyan, TBX18 in magenta and MESP2 in yellow of axioloids embedded in +MG +RAL supplemented with BMS493 at f, 96 h and h, 120 h of culture or supplemented with k, AGN193109 or m, ER50891 at 120 h of culture. g, and i Corresponding quantification of signal intensities along the posterior to anterior axis for +MG +RAL axioloids treated with BMS493 at g, 96 h (2 independent experiments, n = 9) and i, 120 h (4 independent experiments, n = 8) of culture. Lines in e,g and i correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Scale bar is 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 9 Single cell RNA-seq based assessment of RA signaling effect on MG embedded human axioloids.

a, Integrated UMAP projection of single-cell transcriptome profiles of axioloids with all the four conditions (control (no MG), MG only, +MG +RAL (retinal) and +MG +RA (retinoic acid) at both 96 h and 120 h. b, and c, Expression changes by +MG +RAL (retinal) and +MG +RA (retinoic acid) compared to the MG only conditions for consistently up- or down-regulated genes at both 96 h and 120 h in SM b, and TB c. d, Expression changes in SM compared to the control (without MG) samples are indicated for the different conditions (MG only, RAL or RA addition) at both 96 h and 120 h. Genes indicated here are DEGs identified in Fig. 3g (SM 96 h +/–MG). e–g, Expression levels of indicated genes across different conditions in SM for both 96 h and 120 h human axioloids.

Source Data

Extended Data Fig. 10 Morphometric comparison of human axioloids with human CS9, CS10 and CS11 embryos.

a, and b, 3D projection image and image stack-based 3D model creation and somite volume measurement of a CS9 and a CS10 human embryo (left and right-side n = 2, human embryo data obtained from the Virtual Embryo Project https://www.ehd.org/virtual-human-embryo/). c, OPT stack single image and image stack-based 3D model creation and somite volume measurement of the 8 posterior somites of a CS11 human embryo (left and right-side n = 2, OPT data of human embryo obtained from the Human Developmental Biology Resource (HDBR)). d–f, Phalloidin staining and z-stack image-based 3D model creation and somite volume measurement of 409B2 human axioloids embedded in d, +MG +ROL at 120 h (2 independent experiments, n = 4 axioloids), e, +MG +RAL at 120 h (2 independent experiments, n = 4 axioloids), f, +MG +RA at 120 h (2 independent experiments, n = 3 axioloids). g–j, comparison of g, volume, h, area, i, roundness and j, elongation of the somites of 409B2 and 201B7 Luc-derived axioloids cultured in +MG +ROL (n = 32 somites for each cell line), +MG +RAL (n = 32 somites for each cell line) or +MG +RA (409B2 n = 24 somites, 201B7 n = 32 somites) and CS9-VHE (n = 8 somites), CS10-VHE (n = 14 somites) and CS11-OPT (n = 16 somites) human embryos. h, Comparison of the somite area extrapolated from the volumes shown in g with the somite area measured on serial histological sections of CS10 (1 embryo n = 16 somites) and CS11 (3 embryos n = 17 somites) embryos. Boxes in d–f and h–j represent the first and last 4th percentile intervals, whiskers show the minimum and maximum values within that range, middle line the median, and dots represent individual data points. a.u., arbitrary units.

Source Data

Extended Data Fig. 11 Molecular (scRNA-seq) comparison of human axioloids with human CS12 embryo.

a–g, UMAP projection of scRNA-seq datasets of CS12 human embryo. a–b and j–l, Colored by cell types. Annotations are mostly by Xu et al39 except for PSM. b–d, UMAP projection of the integrated scRNA-seq datasets of CS12 human embryo and human axioloids embedded in Matrigel and exposed to retinal. c, Anatomical origin of embryonic samples (rostral and caudal halves of the embryo) are indicated in green (rostral) or blue (caudal) respectively and the TB population found in axioloids is highlighted and shown to match with the caudal (lower/tail) aspect of the embryo. d, The cells from axioloid samples at 96 h or 120 h are highlighted in red (96 h) or green (120 h) on the UMAP projection. The same color code was used to represent in e, the PSM and in g, the EC-like and endothelial cell clusters of axioloid and embryo origin. f, The TB of axioloids (red), and neuromesodermal progenitors (NMP) (green) and neural progenitors (NP) (blue) of the human CS12 embryo are highlighted. h,i, Averaged expression levels of indicated genes in each cluster. h, Shown genes are marker genes of annotated clusters of the CS12 human embryo by Xu et. al and i, marker genes of human axioloids. j, Shows UMAP projection of a redefined PSM cell population based on new marker gene expression, as well as NMP and somitic cells based on the Xu et al. annotation. k, Cells of the CS12 embryo or, l, only the somite, PSM, NMP, and neural progenitor (NP) cells were subjected to velocity analysis and plotted with the same UMAP coordinates as in a and b. RNA velocity is indicated by arrows. m, Expression of TBXT, SOX2, TBX6, HES7, RIPPLY2 and MESP2 genes in the CS12 human embryo on the UMAP projection with magnified views of areas marked by dotted lines shown in the upper right corner. n, Integrated UMAP plot of the axioloid and embryo data. Number of cells from axioloids and from the human embryo origin assigned to each of the 55 identified clusters was used to calculate the Pearson correlation efficiency, results are plotted Fig. 5d.

Extended Data Fig. 12 Molecular characterization of angioblast/EC-like cells found in MG treated human axioloids.

a, Heatmap showing the expression levels of gene markers specific to murine trunk-like structure (TLS) endothelial cells3 or mouse embryo angioblasts56 in the different subpopulations of cells present in human axioloids at 48 h, 72 h, 96 h (+MG), and 120 h (+MG). b, Representative HCR staining with DAPI in gray, TCF15 in yellow, ETV2 in cyan and KDR in magenta of human 201B7 Luc-derived axioloid at 96 h in MG only (top) or +MG +RAL (bottom). Right panel is a magnified view of the rostral aspect of the axioloid. c–d, Z-stack visualization of an immunofluorescence stained 201B7 Luc-derived axioloid in +MG +RAL at 96 h with F-actin (Phalloidin) in gray, MEOX1 in cyan, ETV2 in yellow and KDR in magenta; c, corresponds to the max intensity projection along the z-axis for the MEOX1, ETV2 and KDR channels; d, corresponds to the optical cross section of a somite, which relative position is marked by dashed line on the single plane merged image. Z-stack has been denoised using an AI based software. e–f, High magnification imaging of somite surface with F-actin (Phalloidin) in blue, MEOX1 in gray, ETV2 in yellow and KDR in magenta. In d–f red arrowheads pinpoint to MEOX1-/KDR+/ETV2+ cells, white ones to MEOX1low/KDR+/ETV2+ cells, green one to MEOX1+/KDR-/ETV2+ and blue one to MEOX1+/KDR+/ETV2- cells. Images shown in b are representative of 2 independent experiments and images shown in c–f of 3 independent experiments. Scale bar is 50 µm in d (right panels) and e–f, otherwise 200 µm.

Extended Data Fig. 13 Assessment of the Hox code in human axioloids.

a–d, Pseudotime representation of expression of HOXA, HOXB, HOXC and HOXD cluster associated genes in human MG exposed axioloids at 96 h of culture (24 h after embedding into MG); gene expression patterns arranged along pseudotime rank. e,g, and i, Top panels, analysis of the epigenetic landscape at the HOXA, HOXB and HOXD loci in axioloids at 96 h (24 h after embedding into MG and exposure to retinal) profiled by CUT&Tag using antibodies against H3K4me3 (green) and H3K27me3 (red). Bottom panels, visualization of the spatial distribution of the HOXA, HOXB and HOXD transcripts using HybISS analysis of all the members of the respective HOX clusters using sections of 96 h and 120 h axioloids cultured in +MG +RAL. f,h, and j, Heatmap plots of the HybISS data shown in e,g and i showing the average HOXA, HOXB and HOXD cluster gene expression along the posterior to anterior axis of 96 h and 120 h +MG +RAL axioloids normalized to the position of the MESP2 signal peak (n = 3 axioloids). k, Averaged expression levels of HOXA, HOXB, HOXC, and HOXD genes in whole axioloids exposed to different conditions at 72 h, 96 h and 120 h.

Source Data

Extended Data Fig. 14 Modulating RA, FGF, WNT and Notch signaling in human axioloids.

a–o and s, Quantification of the total HES7:luciferase signal over time and corresponding period measurements as the time interval between the consecutive HES7:Luciferase signal peaks in axioloids +/- MG in a,b, +ROL (retinol) ( 4 independent experiments n = 12 and 5 independent experiments, n = 20), c,d, +RAL (retinal) (5 independent experiments, n = 17 and 5 independent experiments, n = 20), e,f, +RA (retinoic acid) with e, (5 independent experiments, n = 20 and 5 independent experiments, n = 20) and f, (6 independent experiments, n = 22 and 5 independent experiments, n = 20). g–i, Kymographs of the HES7:luciferase signal for axioloids embedded in g, +MG +ROL (3 independent experiments, n = 9), h, +MG +RAL (5 independent experiments, n = 9), and i, +MG +RA (4 independent experiments, n = 8). j, Quantification of average signals in g,h, and i. k, Periodicity measurement of oscillatory activity in g,h and i. l, and m, Kymographs of the HES7:luciferase signal for axioloids embedded in l, +MG +RAL +DMSO (3 independent experiments, n = 6) and m, +MG +RAL +BMS493 (3 independent experiments, n = 5). n, Quantification of average signals in l and m. o, Periodicity measurement of oscillatory activity in l and m. p, Measurement of the total HES7:luciferase signal over time in +MG only (most left) axioloids treated with BMS493 (4 independent experiments, n = 12) and +MG +RAL axioloids treated with BMS493 (3 independent experiments, n = 9), AGN193101 (3 independent experiments, n = 9) and ER50891 (3 independent experiments, n = 9) (most right). q,r, Measurement of the total HES7:luciferase signal over time in +MG +RAL axioloids treated with q, two FGF signaling inhibitors (from left to right) PD173074 (3 independent experiments, n = 9), PD0325901 (5 independent experiments, n = 20), and r, two WNT inhibitors (from left to right) XAV939 (3 independent experiments, n = 9) and IWP2 (5 independent experiments, n = 20). s, Measurement of the total HES7:luciferase signal over time in +MG +RAL axioloids treated with Notch inhibitor DAPT (3 independent experiments, n = 9). t, Representative brightfield images of axioloids embedded in +MG +RAL supplemented with (from left to right) DMSO, DAPT, PD173074, PD0325901, XAV939 and IWP2 at 96 h and 120 h of culture. u, Comparison of the total number of somites (upper panel) and total length (lower panel) of axioloids embedded in +MG +RAL and supplemented with (from left to right) DMSO (3 independent experiments, n = 9), DAPT (3 independent experiments, n = 9), PD173074 (3 independent experiments, n = 9), PD0325901 (3 independent experiments, n = 10), XAV939 (3 independent experiments, n = 9) and IWP2 (3 independent experiments, n = 11) (each reagent is shown in a different colour). Plots in b,d,f,k and o correspond to mean ±SD. Lines in a,c,e,j,n,p,q and s correspond to the mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Boxes in u represent the first and last 4th percentile intervals, whiskers show the minimum and maximum values within that range, middle line the median, and dots represent individual data points. Scale bar is 200 µm. a.u., arbitrary units; n = number of axioloids.

Source Data

Extended Data Fig. 15 Morphological, molecular and functional characterization of patient-like iPSC-derived axioloids with mutations in HES7 and MESP2.

Panels a–e, show data for HES7 KO2, panels f–j, show data for HES7R25W Mt2 and panels k–o, show data for MESP2 KO2; all control and patient-like axioloids were grown under +MG +RAL conditions. a,f, and k, Serial brightfield images of a forming axioloid at 72 h, 96 h and 120 h. b,g, and l, Immunofluorescence staining and signal quantification of +MG +RAL embedded axioloids at 120 h. Top, merged channel images of axioloids stained for F-actin (Phalloidin) in gray, TBXT (BRA) in cyan, FN1 in yellow, and MEOX1 in red. Bottom, corresponding quantification along the posterior to anterior axis of TBXT (blue line) and MEOX1 (in red) signal intensity for HES7 KO2 (3 independent experiments, n = 9), HES7R25W MT2 (3 independent experiments, n = 9) and MESP2 KO2 (3 independent experiments, n = 12). c,h, and m, HCR staining and signal quantification of +MG +RAL embedded axioloids at 120 h, UNCX in cyan and TBX18 in magenta, MESP2 in yellow, and corresponding signal intensity measurements along the posterior to anterior axis normalized to the position of the MESP2 signal peak for HES7 KO2 (3 independent experiments, n = 8), HES7R25W Mt2 (3 independent experiments, n = 8) and MESP2 KO2 (3 independent experiments, n = 9). d,i, and n, Kymograph of HES7 oscillatory activity for d, HES7 KO1 and KO2, i, HES7R25W Mt1 and Mt2, and n, MESP2 KO1 and KO2 patient-like axioloids embedded in +MG +RAL. All experiments were performed under +MG +RAL condition. e,j,o, Average measured HES7:Luciferase signal over time for HES7 KO1 (5 independent experiments, n = 8), HES7 KO2 (6 independent experiments, n = 9), HES7R25W Mt1 (5 independent experiments, n = 10), HES7R25W Mt2 (5 independent experiments, n = 9), MESP2 KO1 (3 independent experiments, n = 6), MESP2 KO2 (3 independent experiments, n = 7) and 201B7 Luc +MG +RAL (5 independent experiments, n = 9 data identical to the one shown in Fig. 14j). Lines in b,c,e,g,h,j,l,m and o correspond to mean values, error bands represent the 95% confidence interval, of which the top and bottom show the 2.5 and 97.5 percentiles for each data point. Scale bar is 200 μm. a.u., arbitrary units; n = number of axioloids.

Source Data

Supplementary information

Supplementary Information

This file contains the Supplementary Discussion, Supplementary Table legends, Supplementary Video legends and Supplementary Figures with legends.

Reporting Summary

Supplementary Table 1

Genes expressed in identified clusters of scRNA-seq data.

Supplementary Table 2

NMP and NE module score genes. List of neuromesodermal progenitor (NMP) and neuroectoderm (NE) associated genes used to calculate the module scores for Extended Data Fig. 3g.

Supplementary Table 3

Recombinant proteins, small molecules, reagents & consumables/equipment used in this study. List of utilized recombinant proteins & small molecules (3.1), modulators of signaling pathways (3.2) reagents (3.3) and consumables/equipment (3.4).

Supplementary Table 4

Antibodies used in this study. List of utilized primary antibodies for immunohistochemistry (4.1), secondary antibodies for immunohistochemistry (4.2), primary antibodies for CUT&Tag library preparation (4.3), secondary antibodies for CUT&Tag library preparation (4.4).

Supplementary Table 5

List of probes used for HCR. List of utilized probes for HCR-based whole-mount in situ hybridization analysis of human axioloid samples.

Supplementary Table 6

List of probe sequences used for HybISS analysis. List of utilized probe sequences used for HybISS-based spatial transcriptomic analysis of human axioloid samples.

Supplementary Table 7

List of primers used for CUT&Tag experiments. List of utilized primers used for CUT&Tag-based analysis of human axioloids.

Supplementary Data

This file contains source data for Supplementary Figs. 1–5.

Supplementary Video 1

Symmetry breaking and initial elongation of axioloids. Live imaging of axioloids derived from 409B2 (upper) and 201B7 Luc (lower) between 24 h and 72 h of culture. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 2

Effect of Matrigel on axioloid morphology. Live imaging of axioloids derived from 409B2 (upper) and 201B7 Luc (lower) with +MG (right) or without -MG (left) embedding into Matrigel (MG) between 72 h and 120 h of culture. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 3

HES7 gene expression dynamics in MG embedded axioloids. Live imaging of the spatiotemporal morphogenetic expression of the HES7 gene in a 201B7 Luc-derived axioloid embedded in MG from 72 h to 120 h of culture. BF video (left) and HES7:Luciferase signal (right). Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 4

Effect of retinoid signaling on axioloid growth and morphology. Live imaging of axioloids derived from 409B2 (top panels) and 201B7 Luc (bottom panels) after embedding in MG only (left) or in MG supplemented with ROL, RAL or RA (right) from 72 h to 120 h of culture. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 5

3D reconstruction of retinoid-treated axioloids. 3D reconstruction of 409B2-derived axioloids embedded in MG only (upper left) or in MG supplemented with ROL (lower left), RAL (lower right) or RA (upper right) at 120 h of culture stained for F-actin (Phalloidin) in gray, TBXT (BRA) in blue, Fibronectin (FN1) in green and MEOX1 in red.

Supplementary Video 6

Effect of RA pathway inhibition on axioloid growth and morphology. Live imaging of axioloids derived from 201B7 Luc after embedding in +MG +RAL supplemented with (from left to right) DMSO, BMS493, AGN193109 or ER50891 from 72 h to 120 h of culture. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 7

3D visualization of axioloids treated with RA pathway inhibitors. 3D reconstruction of axioloids derived from 201B7 Luc embedded in +MG +RAL (upper left) or in +MG +RAL supplemented with three different RA inhibitors, including BMS493 (upper right), AGN193109 (lower left) or ER50891 (lower right) at 120 h of culture stained for F-actin (Phalloidin) in gray, TBXT (BRA) in blue, Fibronectin (FN1) in green and MEOX1 in red.

Supplementary Video 8

Midline and bilateral somite formation in axioloids. Visualization of midline formation in a 409B2-derived axioloid embedded in +MG +RAL (top) and of formation of a single bilateral somite in 409B2-derived axioloid embedded in +MG +ROL (bottom). Live imaging was performed between 72 and 120 h of culture. Scale bar is 200 µm.

Supplementary Video 9

3D reconstruction of somites in axioloids and human CS9-11 embryos. 3D reconstruction of somites formed in 409B2-derived MG embedded axioloids (top) treated ROL (left), RAL (middle) and RA; and 3D reconstructions of somites in human embryos (bottom) found in CS9 (left), CS10 (middle) and CS11 (right) human embryos. Each somite-like structure is highlighted by a different color depending on its position along the antero-posterior axis.

Supplementary Video 10

Effect of ROL, RAL or RA on HES7 gene expression dynamics. Live imaging of the spatiotemporal morphogenetic expression of the HES7 gene in 201B7 Luc-derived axioloids embedded in MG only (upper left pair) or in MG supplemented with ROL (lower left pair), RAL (lower right pair) or RA (upper right pair) from 72 h to 120 h of culture. BF video (left) and HES7:Luciferase signal (right) for each condition. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 11

Effect of BMS493 supplementation on HES7 gene expression dynamics. Live imaging of the spatiotemporal morphogenetic expression of the HES7 gene in 201B7 Luc-derived axioloids embedded in MG supplemented with +RAL+DMSO (left pair) or +RAL+BMS493 (right pair), from 72 h to 120 h of culture. BF video (left) and HES7:Luciferase signal (right) for each condition. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 12

Effect of NOTCH, FGF and WNT pathway inhibition on axioloids. Live imaging of axioloids derived from 201B7 Luc after embedding in +MG +RAL supplemented with DMSO (top left), DAPT (top right), PD173074 (middle left), PD0325901 (middle right), XAV939 (bottom left) or IWP2 (bottom right) from 72 h to 120 h of culture. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 13

Effect of HES7 gene KO on axioloid growth and morphology. Top panel, live imaging of axioloids embedded in +MG +RAL derived from 201B7 Luc (top left), HES7 KO1 (top middle) and HES7 KO2 (top right) cell lines. Data shown is representative of at least three independent experiments. Scale bar is 200 µm. Bottom panel, 3D reconstruction of axioloids embedded in +MG +RAL derived from HES7 KO1 (middle) and HES7 KO2 (bottom) stained with F-actin (Phalloidin) in gray, TBXT (BRA) in blue, Fibronectin (FN1) in green and MEOX1 in red.

Supplementary Video 14

Effect of HES7 gene KO on HES7 gene expression dynamics. Live imaging of the spatiotemporal expression of the HES7 gene in 201B7 Luc, (top pair), HES7 KO1 (middle pair) and HES7 KO2 (bottom pair)-derived axioloids embedded in +MG +RAL. BF video (left) and HES7:Luciferase signal (right) for each condition. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 15

Effect of HES7 point mutation (rs113994160: c.73C>T; HES7R25W) on axioloid growth and morphology. Top panel, live imaging of axioloids embedded in +MG +RAL derived from 201B7 Luc (top left), HES7R25W MT1 (top middle) and HES7R25W MT2 (top right) cell lines. Data shown is representative of at least three independent experiments. Scale bar is 200 µm. Bottom panel, 3D reconstruction of axioloids embedded in +MG +RAL derived from HES7R25W MT1 (middle) and HES7R25W MT2 (bottom) stained with F-actin (Phalloidin) in gray, TBXT (BRA) in blue, Fibronectin (FN1) in green and MEOX1 in red.

Supplementary Video 16

Effect of HES7 point mutation (rs113994160: c.73C>T; HES7R25W) on HES7 gene expression dynamics. Live imaging of the spatiotemporal expression of the HES7 gene in 201B7 Luc, (top pair), HES7R25W MT1 (middle pair) and HES7R25W MT2 (bottom pair)-derived axioloids. BF video (left) and HES7:Luciferase signal (right) for each condition. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

Supplementary Video 17

Effect of MESP2 gene KO on axioloid growth and morphology. Top panel, live imaging of axioloids embedded in +MG +RAL derived from 201B7 Luc (top left), MESP2 KO1 (top middle) and MESP2 KO2 (top right) cell lines. Data shown is representative of at least three independent experiments. Scale bar is 200 µm. Bottom panel, 3D reconstruction of axioloids embedded in +MG +RAL derived from MESP2 KO1 (middle) and MESP2 KO2 (bottom) stained with F-actin (Phalloidin) in gray, TBXT (BRA) in blue, Fibronectin (FN1) in green and MEOX1 in red.

Supplementary Video 18

Effect of MESP2 gene KO on HES7 gene expression dynamics. Live imaging of the spatiotemporal expression of the HES7 gene in 201B7 Luc (top pair), MESP2 KO1 (middle pair) and MESP2 KO2 (bottom pair)-derived axioloids embedded in +MG +RAL. BF video (left) and HES7:Luciferase signal (right) for each condition. Data shown is representative of at least three independent experiments. Scale bar is 200 µm.

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

Yamanaka, Y., Hamidi, S., Yoshioka-Kobayashi, K. et al. Reconstituting human somitogenesis in vitro. Nature 614, 509–520 (2023). https://doi.org/10.1038/s41586-022-05649-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41586-022-05649-2

This article is cited by

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

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