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.

Multilineage communication regulates human liver bud development from pluripotency

Abstract

Conventional two-dimensional differentiation from pluripotency fails to recapitulate cell interactions occurring during organogenesis. Three-dimensional organoids generate complex organ-like tissues1; however, it is unclear how heterotypic interactions affect lineage identity. Here we use single-cell RNA sequencing2,3 to reconstruct hepatocyte-like lineage progression from pluripotency in two-dimensional culture. We then derive three-dimensional liver bud organoids4 by reconstituting hepatic, stromal, and endothelial interactions, and deconstruct heterogeneity during liver bud development. We find that liver bud hepatoblasts diverge from the two-dimensional lineage, and express epithelial migration signatures characteristic of organ budding. We benchmark three-dimensional liver buds against fetal and adult human liver single-cell RNA sequencing data, and find a striking correspondence between the three-dimensional liver bud and fetal liver cells. We use a receptor–ligand pairing analysis and a high-throughput inhibitor assay to interrogate signalling in liver buds, and show that vascular endothelial growth factor (VEGF) crosstalk potentiates endothelial network formation and hepatoblast differentiation. Our molecular dissection reveals interlineage communication regulating organoid development, and illuminates previously inaccessible aspects of human liver development.

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

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

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

Figure 1: Exploring human hepatic differentiation from pluripotency in 2D culture and 3D LB organoids.
Figure 2: Hypoxia and extracellular matrix modulation in the LB microenvironment induce hepatic outgrowth signatures and prime LBs for vascularization.
Figure 3: A divergent hepatic lineage trajectory in the LB generates cells that resemble fetal human hepatocytes.
Figure 4: Interlineage signalling affects LB development.

Accession codes

Primary accessions

Gene Expression Omnibus

References

  1. Lancaster, M. A. & Knoblich, J. A. Organogenesis in a dish: modeling development and disease using organoid technologies. Science 345, 1247125 (2014)

    PubMed  Google Scholar 

  2. Treutlein, B. et al. Reconstructing lineage hierarchies of the distal lung epithelium using single-cell RNA-seq. Nature 509, 371–375 (2014)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  3. Camp, J. G. et al. Human cerebral organoids recapitulate gene expression programs of fetal neocortex development. Proc. Natl Acad. Sci. USA 112, 15672–15677 (2015)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  4. Takebe, T. et al. Vascularized and complex organ buds from diverse tissues via mesenchymal cell-driven condensation. Cell Stem Cell 16, 556–565 (2015)

    CAS  PubMed  Google Scholar 

  5. Takebe, T. et al. Vascularized and functional human liver from an iPSC-derived organ bud transplant. Nature 499, 481–484 (2013)

    ADS  CAS  PubMed  Google Scholar 

  6. Rankin, S. A., Kormish, J., Kofron, M., Jegga, A. & Zorn, A. M. A gene regulatory network controlling hhex transcription in the anterior endoderm of the organizer. Dev. Biol. 351, 297–310 (2011)

    CAS  PubMed  PubMed Central  Google Scholar 

  7. Teo, A. K. et al. Pluripotency factors regulate definitive endoderm specification through eomesodermin. Genes Dev. 25, 238–250 (2011)

    CAS  PubMed  PubMed Central  Google Scholar 

  8. Si-Tayeb, K. et al. Highly efficient generation of human hepatocyte-like cells from induced pluripotent stem cells. Hepatology 51, 297–305 (2010)

    CAS  PubMed  Google Scholar 

  9. Zaret, K. S. Regulatory phases of early liver development: paradigms of organogenesis. Nat. Rev. Genet. 3, 499–512 (2002)

    CAS  PubMed  Google Scholar 

  10. Sosa-Pineda, B., Wigle, J. T. & Oliver, G. Hepatocyte migration during liver development requires Prox1. Nat. Genet. 25, 254–255 (2000)

    CAS  PubMed  Google Scholar 

  11. Margagliotti, S. et al. The Onecut transcription factors HNF-6/OC-1 and OC-2 regulate early liver expansion by controlling hepatoblast migration. Dev. Biol. 311, 579–589 (2007)

    CAS  PubMed  Google Scholar 

  12. Ding, B. S. et al. Inductive angiocrine signals from sinusoidal endothelium are required for liver regeneration. Nature 468, 310–315 (2010)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  13. Hu, J. et al. Endothelial cell-derived angiopoietin-2 controls liver regeneration as a spatiotemporal rheostat. Science 343, 416–419 (2014)

    ADS  CAS  PubMed  Google Scholar 

  14. Zorn, A. M. in StemBook (2008)

  15. Beg, A. A., Sha, W. C., Bronson, R. T., Ghosh, S. & Baltimore, D. Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-κB. Nature 376, 167–170 (1995)

    ADS  CAS  PubMed  Google Scholar 

  16. Kamiya, A. & Gonzalez, F. J. TNF regulates mouse fetal hepatic maturation induced by oncostatin M and extracellular matrices. Hepatology 40, 527–536 (2004)

    CAS  PubMed  Google Scholar 

  17. Carmeliet, P. & Jain, R. K. Molecular mechanisms and clinical applications of angiogenesis. Nature 473, 298–307 (2011)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  18. D’Amico, G. et al. Tie1 deletion inhibits tumor growth and improves angiopoietin antagonist therapy. J. Clin. Invest. 124, 824–834 (2014)

    PubMed  PubMed Central  Google Scholar 

  19. Vollmar, B. & Menger, M. D. The hepatic microcirculation: mechanistic contributions and therapeutic targets in liver injury and repair. Physiol. Rev. 89, 1269–1339 (2009)

    CAS  PubMed  Google Scholar 

  20. Dunwoodie, S. L. The role of hypoxia in development of the mammalian embryo. Dev. Cell 17, 755–773 (2009)

    CAS  PubMed  Google Scholar 

  21. Matsumoto, K., Yoshitomi, H., Rossant, J. & Zaret, K. S. Liver organogenesis promoted by endothelial cells prior to vascular function. Science 294, 559–563 (2001)

    ADS  CAS  PubMed  Google Scholar 

  22. Takayama, N. et al. Transient activation of c-MYC expression is critical for efficient platelet generation from human induced pluripotent stem cells. J. Exp. Med. 207, 2817–2830 (2010)

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Loh, K. M. et al. Efficient endoderm induction from human pluripotent stem cells by logically directing signals controlling lineage bifurcations. Cell Stem Cell 14, 237–252 (2014)

    CAS  PubMed  PubMed Central  Google Scholar 

  24. Kajiwara, M. et al. Donor-dependent variations in hepatic differentiation from human-induced pluripotent stem cells. Proc. Natl Acad. Sci. USA 109, 12538–12543 (2012)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  25. Takebe, T. et al. Generation of a vascularized and functional human liver from an iPSC-derived organ bud transplant. Nat. Protocols 9, 396–409 (2014)

    CAS  PubMed  Google Scholar 

  26. Kegel, V. et al. Protocol for isolation of primary human hepatocytes and corresponding major populations of non-parenchymal liver cells. J. Vis. Exp. 109, e53069 (2016)

    Google Scholar 

  27. Lancaster, M. A. & Knoblich, J. A. Generation of cerebral organoids from human pluripotent stem cells. Nat. Protocols 9, 2329–2340 (2014)

    CAS  PubMed  Google Scholar 

  28. Renaud, G., Kircher, M., Stenzel, U. & Kelso, J. freeIbis: an efficient basecaller with calibrated quality scores for Illumina sequencers. Bioinformatics 29, 1208–1209 (2013)

    CAS  PubMed  PubMed Central  Google Scholar 

  29. Renaud, G., Stenzel, U., Maricic, T., Wiebe, V. & Kelso, J. deML: robust demultiplexing of Illumina sequences using a likelihood-based approach. Bioinformatics 31, 770–772 (2015)

    CAS  PubMed  Google Scholar 

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

    CAS  PubMed  PubMed Central  Google Scholar 

  31. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009)

    CAS  PubMed  PubMed Central  Google Scholar 

  32. Trapnell, C. et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515 (2010)

    CAS  PubMed  PubMed Central  Google Scholar 

  33. Macosko, E. Z. et al. Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Cell 161, 1202–1214 (2015)

    CAS  PubMed  PubMed Central  Google Scholar 

  34. 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)

    CAS  PubMed  PubMed Central  Google Scholar 

  35. Huang, W ., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protocols 4, 44–57 (2008)

    Google Scholar 

  36. Zhang, H. M. et al. AnimalTFDB: a comprehensive animal transcription factor database. Nucleic Acids Res. 40, D144–D149 (2012)

    CAS  PubMed  Google Scholar 

  37. Kim, D. H. et al. Single-cell transcriptome analysis reveals dynamic changes in lncRNA expression during reprogramming. Cell Stem Cell 16, 88–101 (2015)

    CAS  PubMed  PubMed Central  Google Scholar 

  38. Wirth, H., Löffler, M., von Bergen, M. & Binder, H. Expression cartography of human tissues using self organizing maps. BMC Bioinformatics 12, 306 (2011)

    PubMed  PubMed Central  Google Scholar 

  39. Wirth, H., von Bergen, M. & Binder, H. Mining SOM expression portraits: feature selection and integrating concepts of molecular function. BioData Min. 5, 18 (2012)

    PubMed  PubMed Central  Google Scholar 

  40. Ashburner, M. et al.; The Gene Ontology Consortium. Gene Ontology: tool for the unification of biology. Nat. Genet. 25, 25–29 (2000)

    CAS  PubMed  PubMed Central  Google Scholar 

  41. Liberzon, A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst. 1, 417–425 (2015)

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Törönen, P., Ojala, P. J., Marttinen, P. & Holm, L. Robust extraction of functional signals from gene set analysis using a generalized threshold free scoring function. BMC Bioinformatics 10, 307 (2009)

    PubMed  PubMed Central  Google Scholar 

  43. Strimmer, K. fdrtool: a versatile R package for estimating local and tail area-based false discovery rates. Bioinformatics 24, 1461–1462 (2008)

    CAS  PubMed  Google Scholar 

  44. Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005)

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  45. Ramilowski, J. A. et al. A draft network of ligand-receptor-mediated multicellular signalling in human. Nat. Commun. 6, 7866 (2015)

    ADS  CAS  PubMed  Google Scholar 

  46. Rodríguez-Seguel, E. et al. Mutually exclusive signaling signatures define the hepatic and pancreatic progenitor cell lineage divergence. Genes Dev. 27, 1932–1946 (2013)

    PubMed  PubMed Central  Google Scholar 

  47. Scialdone, A. et al. Computational assignment of cell-cycle stage from single-cell transcriptome data. Methods 85, 54–61 (2015)

    CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank S. Pääbo, K. Köhler, A. Weigert, B. Höber, A. Weihmann, B. Vernot, K. Pfüfer, G. Renaud, M. Dannemann, J. Kelso of Max Planck Institute for Evolutionary Anthropology, M. Zerial of the Max Planck Institute of Molecular Cell Biology and Genetics, and T. Tamura, S. Tsuzuki, and C. Kaneko of Yokohama for their help with this project. This work was supported by the Max Planck Society (B.T.), PRESTO Japan Science and Technology Agency (T.T.), grants-in-aid 15H04922 and 15KK0314 from the Ministry of Education Culture and Sports of Japan (K.S.), and the AMED Research Center Network for Realization of Regenerative Medicine (H.T.). T.T. is a New York Stem Cell Foundation – Robertson Investigator.

Author information

Authors and Affiliations

Authors

Contributions

T.T. and B.T. contributed equally to this work. J.G.C., K.S., B.T., and T.T. designed the study and wrote the manuscript. J.G.C. and K.S. performed single-cell experiments with assistance from T.G., S.K., B.T. and M.G. K.S., E.Y., M.K., and H.A. performed LB siRNA experiments, transplants, and stainings. R.O. performed fluorescence-activated cell sorting. G.D. and D.S. isolated human liver cells. J.G.C., B.T., and L.B. performed mouse liver scRNA-seq. J.G.C., K.S., M.B., and R.B. performed high-throughput inhibition experiments, and J.K. and J.G.C. analysed image data. H.L.W. and H.B. assisted with SOM analysis. H.T. provided financial and intellectual support.

Corresponding authors

Correspondence to Keisuke Sekine, Takanori Takebe or Barbara Treutlein.

Ethics declarations

Competing interests

T.T. and H.T. have served on a scientific advisory board for Healios Inc., and granted a licence to Healios through Yokohama City University over inventions related to this manuscript.

Additional information

Reviewer Information Nature thanks S. Linnarsson and the other anonymous reviewer(s) for their contribution to the peer review of this work.

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 Figure 1 Characterizing hepatocyte-like lineage progression.

a, Schematic showing time points for scRNA-seq experiments performed on iPS, DE, HE, IH, and MH cells during 2D hepatogenesis. b, Hierarchical clustering was performed on genes identified by PCA and is shown in e. Intermixing of cells from later time points with earlier time points was used to calculate the efficiency of lineage progression shown in c. c, Lineage progression efficiency at each time point shows the percentage of cells that cluster with cells from previous time points (grey). d, Monocle reveals hepatocyte-like lineage progression. Cells (circles, coloured on the basis of time point) are arranged in the 2D independent component space on the basis of genes identified through PCA (Supplementary Table 1). The minimum spanning tree (grey lines) connects cells, with the black line indicating the longest path. e, Ordering of scRNA-seq expression data according to the pseudotemporal position along the lineage reveals a continuum of gene expression changes from pluripotency to hepatocyte-like fate. Each column represents a single cell, each row a gene. The time point is shown as sidebar. f, Correlating cells with microdissected E8.5 ventral foregut (orange) or E10.5 ventral pancreas (dark grey), dorsal pancreas (light grey), or liver (purple) shows that transitions observed in vitro reflect in vivo development. Cells are ordered on the basis of pseudotime. g, Transcription factor correlation network during hepatoblast lineage progression. Shown are nodes (TFs) with more than three edges, with each edge representing a high correlation (>0.3) between connected TFs. For a full list of network TFs, see Supplementary Table 1.

Extended Data Figure 2 Identifying gene expression programs controlling hepatic differentiation in 2D.

a, Normalized expression of three genes shows examples of dynamic gene expression along the hepatocyte lineage. Cells are ordered according to Monocle pseudotime. b, Covariation network using genes that have high correlation (>0.3) with TFs controlling the hepatocyte-like lineage. TF nodes are coloured according to the time point of maximal expression. Nearest neighbour subnetworks for (1) CER1, (2) TBX3, and (3) AFP show genes that strongly correlated with these known DE, HE, and hepatoblast marker genes. See Fig. 1f for base TF network. ch, Violin plots and expression of subnetwork genes in cells ordered by Pseudotome. Barplots show the number of citations for each gene co-occurring with either ‘liver’, ‘hepatogenesis’, or ‘endoderm’ in PubMed abstracts and suggests known and novel genes involved in hepatocyte development. i, Expression of multiple genes with no described role in liver development in mouse E8.5 foregut and E10.5 liver confirms that these genes are dynamically expressed during liver development.

Extended Data Figure 3 Genotyping LB cells using informative SNPs derived from scRNA-seq of input cells.

a, b, iPS cells, ECs, and MCs used in this study were from different humans. Single-cell RNA-seq reads from iPS cells, ECs, and MCs were used to identify informative SNPs that were present in one human but not the others. Each LB cell was then genotyped on the basis of PCA clustering (a) and the fraction of shared SNPs with each input lineage (b). c, PCA separated each input and LB lineage, suggesting that each lineage undergoes distinct transcriptional changes upon co-culture. PCA was performed on all single-cell transcriptomes using genes expressed in more than two cells and with a non-zero variance. d, Boxplots showing average log2(fold-change distributions) for pairwise comparisons between 2D cultures of hepatic ectoderm, ECs, and MCs with their LB counterparts.

Extended Data Figure 4 SOM metagenes reconstruct hepatocyte-like lineage progression.

a, SOM analysis of 418 hepatic lineage cells using 12,150 genes organized into 100 metagenes was used to reconstruct lineage relations on the basis of a minimum spanning tree. Shown are representative single-cell portraits as well as the consensus portrait for each time point. Portraits are linearly scaled, amplifying higher expressed metagenes. b, Lineage network constructed from pairwise correlations between cells on the basis of the expression of 100 metagenes shows a differentiation topology from iPS cells to DE–HE–IH–MH cells, similar to Fig. 1b. c, Metagenes were clustered into signatures on the basis of their correlation across the cells. Each cell’s correlation with each signature is shown to the right. d, Gene set enrichment analysis was done using genes composing each metagene signature. Top scoring enrichments for clusters 1–2, clusters 3–4, and clusters 5–7 are shown. Dark orange and blue represent a strong and weak correlation for each cell, respectively. e, Boxplots show gene set enrichment (GSE) distributions for each time point using curated sets of genes with specific expression in liver, cerebral cortex, pancreas, or testes.

Extended Data Figure 5 Exploring metagene signature shifts in LBs.

a, PCA based on the expression of 2,500 metagenes separates each input and LB lineage. b, Population map showing the number of genes located in each metagene. c, Map showing the variance for each metagene across all single cells. d, Bar plot showing the proportion of genes and metagenes that are differentially expressed between input and LB populations. e, Heatmap showing normalized overexpression scores for each cell for each signature. f, Representative single-cell maps are shown for each overexpression signature. Bar plots show the percentage of cells from each group that display significant metagene signature expression. Each metagene signature is annotated with GO enrichments, representative genes, and the number of genes and metagenes constituting the signature. g, Radar plots show enrichments for genes involved in cholesterol metabolism, extracellular matrix, angiogenesis, mitosis, translation, hypoxia, endoplasmic reticulum stress, inflammation, and TNF/NF-κB signalling for each signature. Hallmark gene sets were sourced from MSigDB. h, Dot plot shows the expression of genes differentially expressed in HE-LB, where the size of the circle represents the proportion of cells expressing the gene and the colour represents the average expression level of each gene. i, Violin plot shows the distribution of scores for each EC and EC-LB cell on the basis of the expression of genes upregulated during the angiogenic induction phase of liver sinusoidal ECs after partial hepatectomy13. j, Assignment of cell cycle stage for each EC and EC-LB cell47. Note that 100% of EC-LB cells were scored in the G1 phase of the cell cycle. k, Boxplots show normalized gene set enrichment scores for the Hallmark IL6/JAK/STAT3 signalling pathway for each input and LB cell. Heatmap shows the expression of genes in this Hallmark pathway. Expression is scaled across genes. Top sidebar shows the cell type with the gene set enrichment z-score for each cell above. HE (light red), HE-LB (dark red), EC (dark blue), EC-LB (light blue), MC (dark green), and MC-LB (light green).

Extended Data Figure 6 HE cells within the LB start to mature and express genes involved in hepatic outgrowth.

a, Progressive differentiation of HE cells towards hepatoblast within the LB was confirmed in an intercellular correlation network using all genes used to construct the correlation network (Fig. 1). b, Violin plots showing the distribution of marker gene expression that are upregulated during the transition from HE to IH/MH, confirming that HE-LB cells are at intermediate stages of maturation. c, PCA followed by tSNE shows that many HE-LB cells have gene expression profiles different from cells on the 2D hepatocyte lineage, suggesting that cells have distinct transcriptional responses during LB self-organization. d, Dot plot shows the expression of genes differentially expressed in HE-LB, where the size of the circle represents the proportion of cells expressing the gene and the colour represents the average expression level of each gene. e, ONECUT2 and PROX1, genes involved in hepatic outgrowth, are upregulated specifically in HE-LB and in E10.5 mouse liver.

Extended Data Figure 7 Cell composition in human longer-term LBs, adult and fetal human liver, and mouse embryonic liver.

a, scRNA-seq was performed on LBs grown for an extended period in culture (LB1 = 11 days in 3D, LB2 = 12 days, LB3 = 7 days, LB4 = 13 days, LB5 = 10 days; 173 cells total). PCA was used to identify genes describing cell populations, and tSNE was used to cluster cells. We identified hepatic, endothelial, and MC clusters. b, Expression of genes marking each cluster is coloured (log2(FPKM scale)). c, Scaled heatmap (yellow, high; purple, low) of genes specifically expressed (average difference >2, power to discriminate >0.5) in each cluster (labelled at the bottom), with exemplary genes from each cluster labelled on the right. Cells are in columns, genes in rows. d, scRNA-seq was performed on human adult (5 donors, age 21 – 65, 256 cells) and fetal (gestation weeks 10.5 and 17.5, 238 cells) after enrichment for hepatic lineage cells. PCA was used to identify genes describing cell populations, and tSNE was used to cluster cells. We identified hepatic, endothelial, mesenchymal, and immune lineage cells both in fetal and in adult tissues. e, Cluster assignments are highlighted on the tSNE plot and expression of genes marking each cluster is coloured (log2(FPKM scale)). f, Scaled heatmap (yellow, high; purple, low) of genes marking each cluster (numbered at the bottom), with exemplary genes from each cluster labelled on the right. Cells are in columns, genes in rows. g, Marker genes (average difference >2, power to discriminate cluster >0.5) were analysed for GO enrichments, and representative top enrichments are shown with the P value (−log10). h, scRNA-seq was performed on mouse hepatoblasts isolated from mouse liver at E14.5, E15.5, and E16.5 (92 cells total). PCA was used to identify genes describing cell populations, and tSNE was used to cluster cells. We identified hepatic, mesenchymal, and immune cell clusters. i, Expression of genes marking each cluster is coloured (log2(FPKM scale)). j, Scaled heatmap (yellow, high; purple, low) of genes specifically expressed (average difference >2, power to discriminate >0.5) in each cluster (labelled at the bottom), with exemplary genes from each cluster labelled on the right. Cells are in columns, genes in rows. k, Heatmap showing normalized expression (z-score) of genes differentially expressed between fetal and adult human hepatocytes. Column, gene; row, cell. Heatmap shows that fetal hepatocyte marker genes are highly expressed in mouse AFP+, ALB+, DLK+ early hepatocytes/hepatoblasts, corroborating the human fetal data.

Extended Data Figure 8 LB hepatic cells resemble human fetal hepatocytes.

a, Overview of quadratic programming. Fractional identities are calculated assuming a linear combination of different cell fates. Mock-bulk transcriptomes from DE, fetal hepatocytes, and adult hepatocytes were generated by averaging the single-cell transcriptomes from each group. b, For each cell, the similarity to bulk RNA-seq from either DE or fetal hepatocytes was calculated by using quadratic programming and plotted as fractional identities (left axis, circle, fractional DE identity; right axis, triangle, fractional fetal hepatocyte identity). Points are coloured on the basis of the experiment: DE (orange), HE (red), IH (pink), MH (purple), early LB hepatic (LB-early, 3 days, teal), late LB hepatic (LB-late, 5 – 10 days, light blue), and transplanted hepatic (transplant, 5 – 10 days, dark blue). c, For all cells on the 2D and 3D lineage, quadratic programming was used to calculate the fractional identity of each cell’s transcriptome with mock-bulk transcriptomes from DE or hepatocytes (fetal or adult). Fetal (solid line) and adult (dotted line) hepatocyte fractional identity is plotted for 2D (pink) and 3D (blue) lineage cells ordered in ascending order. Hepatic cells from the LB are the most similar to fetal hepatocytes. d, Heatmap shows normalized correlation (z-score) of single-cell transcriptomes with mock-bulk RNA-seq data from adult hepatocytes, fetal hepatocytes, hepatic cells from late stage LBs, and 2D MH cells. e, Correlogram showing the correlation between cells from different cell populations. The left sidebar shows the expression of ALB, AFP, and DLK1. f, Heatmap showing differentially expressed genes between 2D and 3D lineages for ALB+ cells in MH, late LB hepatic, transplanted hepatic cells (top), and fetal and adult hepatocytes (bottom).

Extended Data Figure 9 Similarities and differences between endothelial and MCs from the LB and primary human liver.

a, For each cell, the similarity to either mock-bulk RNA-seq data from input MSCs grown in 2D culture or primary stellate cells from the fetal liver was calculated using quadratic programming and plotted as fractional identities (left axis, circle, fractional MSC 2D identity; right axis, triangle, fractional stellate identity). Points are coloured on the basis of the experiment: input MSCs in 2D (MSC, light green), MSCs from early LB (MSC-LB early, red), MSCs from late LB (MSC-LB late, purple), primary fetal stellate (fetal stellate, green). b, Heatmap shows normalized correlation (z-score) of single-cell transcriptomes with mock-bulk RNA-seq data from MSCs from early stage LBs, MSCs from late stage LBs, MSCs grown in 2D, and fetal stellate cells. c, tSNE clustering of mesenchymal and stellate cells on the basis of genes identified using PCA. This analysis was used to identify genes specific to fetal stellate cells (Supplementary Table 5). d, Violin plots showing the expression distributions of genes similarly and differentially expressed between MSC-LB and fetal stellate cells. White circle, mean. e, For each cell, the similarity to either mock-bulk RNA-seq data from input ECs grown in 2D culture or primary ECs from fetal and adult liver was calculated using quadratic programming and plotted as fractional identities (left axis, circle, fractional EC 2D identity; right axis, triangle, fractional primary EC identity). Points are coloured on the basis of the experiment: input ECs in 2D (EC, green), ECs from early LB (EC-LB, red), and primary fetal and adult ECs (primary EC, red). f, Heatmap shows normalized correlation (z-score) of single-cell transcriptomes with mock-bulk RNA-seq data from ECs grown in 2D, ECs from early stage LBs, and primary liver ECs. g, tSNE clustering of ECs on the basis of genes identified using PCA. This analysis was used to identify genes specific to primary liver ECs (Supplementary Table 5). h, Violin plots showing the expression distributions of genes similarly and differentially expressed between EC-LB and primary endothelial cells. White circle, mean.

Extended Data Figure 10 Potential interlineage signalling in LBs.

a, b, t-SNE plots based on the expression of receptors and ligands show that input and LB cells cluster separately. c, Heatmap shows the scaled expression of receptors and ligands differentially expressed between HE, EC, and MC cells in the early LB. d, Boxplots show the number of ligands (L) or receptors (R) for each lineage combination. For example, hepatoblasts average 150 ligands and 100 receptors for every HE–EC interaction. e, Violin jitter plots showing the number of potential cell interactions for each lineage combination calculated from the network shown in b. f, Top GO enrichments and P values for receptor–ligand pairings where one of the three cell types (EC, MC, or HE) are required for the interaction. g, Each of the three LB lineages (HE, red; EC, green; MC, blue) are represented as a third of the circle. The fraction of cells expressing mRNA for each ligand is highlighted and linked to the fraction of cells expressing the receptor. The arrows designate the direction of the link (ligand to receptor). A subset of cells may express both receptor and ligand and are shaded accordingly. h, qRT–PCR for hepatoblast marker genes shows that hepatic differentiation is impaired in LBs generated from ECs with knockdown of an endothelial specific receptor, TIE1 (top), or endothelial specific ligand, EDN1 (bottom). LBs were generated containing TIE1 or EDN1 knockdown ECs, combined with normal wild-type HEs and MCs. Two-sided t-test, *P < 0.05; error bars, s.d. Data from three independent differentiations. i, Distributions of mean fluorescence intensities from maximum projections of each micro-LB in the green (hepatic) and red (endothelial) channels. These signals were used to calculate a hepatic-to-endothelial signal ratio for each micro-LB. j, Top: violin plot showing the expression distributions of JAK3 expression across input and early LB cell types. Bottom: representative maximum projection images of a micro-LB at the 48 h time point from a DMSO-treated control and with 10 μM tasocitinib, an inhibitor targeting JAK3 of the JAK/STAT pathway. k, Addition of KDR/VEGFR2 inhibitor (SU1498) does not affect the self-condensation process; however, endothelial sprouting is impaired in the presence of KDR inhibitor. l, qRT–PCR for hepatoblast marker genes shows that hepatic differentiation is not significantly impaired in HE monoculture in the presence of KDR inhibitor. Error bars, s.d. Data from three independent differentiations.

Supplementary information

Supplementary Table 1

Genes, data tables, and ontology enrichments for main and extended data figures. (XLSX 18314 kb)

Supplementary Table 2

Summary of metagene signature from the LB SOM analysis. (XLSX 589 kb)

Supplementary Table 3

Signature scores for each LB cell from the SOM analysis. (XLSX 175 kb)

Supplementary Table 4

Cell composition analysis of late LBs, mouse liver, and human fetal and adult liver. (XLSX 1588 kb)

Supplementary Table 5

Differential gene expression between primary and LB hepatic, mesenchymal, and endothelial cells. (XLSX 78 kb)

Supplementary Table 6

High-throughput inhibitor expression data and metadata. (XLSX 1172 kb)

Supplementary Information

This file contains the full legends for Supplementary Tables 1-6. (PDF 53 kb)

Supplementary Data

This file contains the custom code used to analyse the scRNA-seq data. (ZIP 84328 kb)

PowerPoint slides

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Camp, J., Sekine, K., Gerber, T. et al. Multilineage communication regulates human liver bud development from pluripotency. Nature 546, 533–538 (2017). https://doi.org/10.1038/nature22796

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nature22796

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

Quick links

Nature Briefing: Translational Research

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

Get what matters in translational research, free to your inbox weekly. Sign up for Nature Briefing: Translational Research