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.

Haematopoietic stem cell induction by somite-derived endothelial cells controlled by meox1

Abstract

Haematopoietic stem cells (HSCs) are self-renewing stem cells capable of replenishing all blood lineages. In all vertebrate embryos that have been studied, definitive HSCs are generated initially within the dorsal aorta (DA) of the embryonic vasculature by a series of poorly understood inductive events1,2,3. Previous studies have identified that signalling relayed from adjacent somites coordinates HSC induction, but the nature of this signal has remained elusive4. Here we reveal that somite specification of HSCs occurs via the deployment of a specific endothelial precursor population, which arises within a sub-compartment of the zebrafish somite that we have defined as the endotome. Endothelial cells of the endotome are specified within the nascent somite by the activity of the homeobox gene meox1. Specified endotomal cells consequently migrate and colonize the DA, where they induce HSC formation through the deployment of chemokine signalling activated in these cells during endotome formation. Loss of meox1 activity expands the endotome at the expense of a second somitic cell type, the muscle precursors of the dermomyotomal equivalent in zebrafish, the external cell layer. The resulting increase in endotome-derived cells that migrate to colonize the DA generates a dramatic increase in chemokine-dependent HSC induction. This study reveals the molecular basis for a novel somite lineage restriction mechanism and defines a new paradigm in induction of definitive HSCs.

Your institute does not have access to this article

Relevant articles

Open Access articles citing this article.

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

All prices are NET prices.

Figure 1: cho mutants possess defects in the ECL and the muscles derived from it.
Figure 2: Endotome markers persist in the DA and are expanded in meox1 mutants.
Figure 3: Endotome cells contribute to DA formation and are expanded in meox1 mutants.
Figure 4: Somite-derived endothelial cells are required for HSC induction.

References

  1. Davidson, A. J. & Zon, L. I. The 'definitive' (and 'primitive') guide to zebrafish hematopoiesis. Oncogene 23, 7233–7246 (2004)

    CAS  PubMed  Article  Google Scholar 

  2. Kissa, K. & Herbomel, P. Blood stem cells emerge from aortic endothelium by a novel type of cell transition. Nature 464, 112–115 (2010)

    ADS  CAS  PubMed  Article  Google Scholar 

  3. Bertrand, J. Y. et al. Haematopoietic stem cells derive directly from aortic endothelium during development. Nature 464, 108–111 (2010)

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  4. Clements, W. K. et al. A somitic Wnt16/Notch pathway specifies haematopoietic stem cells. Nature 474, 220–224 (2011)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  5. Noden, D. M. Embryonic origins and assembly of blood vessels. Am. Rev. Respir. Dis. 140, 1097–1103 (1989)

    CAS  PubMed  Article  Google Scholar 

  6. Pardanaud, L. & Dieterlen-Lievre, F. Does the paraxial mesoderm of the avian embryo have hemangioblastic capacity? Anat. Embryol. (Berl.) 192, 301–308 (1995)

    CAS  Article  Google Scholar 

  7. Pardanaud, L. et al. Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis. Development 122, 1363–1371 (1996)

    CAS  PubMed  Article  Google Scholar 

  8. Wilting, J. et al. Angiogenic potential of the avian somite. Dev. Dyn. 202, 165–171 (1995)

    CAS  PubMed  Article  Google Scholar 

  9. Jaffredo, T., Gautier, R., Eichmann, A. & Dieterlen-Lievre, F. Intraaortic hemopoietic cells are derived from endothelial cells during ontogeny. Development 125, 4575–4583 (1998)

    CAS  PubMed  Article  Google Scholar 

  10. Pouget, C., Gautier, R., Teillet, M. A. & Jaffredo, T. Somite-derived cells replace ventral aortic hemangioblasts and provide aortic smooth muscle cells of the trunk. Development 133, 1013–1022 (2006)

    CAS  PubMed  Article  Google Scholar 

  11. Esner, M. et al. Smooth muscle of the dorsal aorta shares a common clonal origin with skeletal muscle of the myotome. Development 133, 737–749 (2006)

    CAS  PubMed  Article  Google Scholar 

  12. Ben-Yair, R. & Kalcheim, C. Notch and bone morphogenetic protein differentially act on dermomyotome cells to generate endothelium, smooth, and striated muscle. J. Cell Biol. 180, 607–618 (2008)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  13. Childs, S., Chen, J. N., Garrity, D. M. & Fishman, M. C. Patterning of angiogenesis in the zebrafish embryo. Development 129, 973–982 (2002)

    CAS  PubMed  Article  Google Scholar 

  14. Jin, S. W., Beis, D., Mitchell, T., Chen, J. N. & Stainier, D. Y. Cellular and molecular analyses of vascular tube and lumen formation in zebrafish. Development 132, 5199–5209 (2005)

    CAS  PubMed  Article  Google Scholar 

  15. Kohli, V., Schumacher, J. A., Desai, S. P., Rehn, K. & Sumanas, S. Arterial and venous progenitors of the major axial vessels originate at distinct locations. Dev. Cell 25, 196–206 (2013)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  16. Lawson, N. D. & Weinstein, B. M. In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev. Biol. 248, 307–318 (2002)

    CAS  PubMed  Article  Google Scholar 

  17. Williams, C. et al. Hedgehog signaling induces arterial endothelial cell formation by repressing venous cell fate. Dev. Biol. 341, 196–204 (2010)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  18. Zhang, X. Y. & Rodaway, A. R. SCL-GFP transgenic zebrafish: in vivo imaging of blood and endothelial development and identification of the initial site of definitive hematopoiesis. Dev. Biol. 307, 179–194 (2007)

    CAS  PubMed  Article  Google Scholar 

  19. Svetic, V. et al. Sdf1a patterns zebrafish melanophores and links the somite and melanophore pattern defects in choker mutants. Development 134, 1011–1022 (2007)

    CAS  PubMed  Article  Google Scholar 

  20. Hollway, G. E. et al. Whole-somite rotation generates muscle progenitor cell compartments in the developing zebrafish embryo. Dev. Cell 12, 207–219 (2007)

    CAS  PubMed  Article  Google Scholar 

  21. Stellabotte, F., Dobbs-McAuliffe, B., Fernandez, D. A., Feng, X. & Devoto, S. H. Dynamic somite cell rearrangements lead to distinct waves of myotome growth. Development 134, 1253–1257 (2007)

    CAS  PubMed  Article  Google Scholar 

  22. Cha, Y. R. et al. Chemokine signaling directs trunk lymphatic network formation along the preexisting blood vasculature. Dev. Cell 22, 824–836 (2012)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  23. Bussmann, J., Bakkers, J. & Schulte-Merker, S. Early endocardial morphogenesis requires Scl/Tal1. PLoS Genet. 3, e140 (2007)

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  24. Yabe, T. & Takada, S. Mesogenin causes embryonic mesoderm progenitors to differentiate during development of zebrafish tail somites. Dev. Biol. 370, 213–222 (2012)

    CAS  PubMed  Article  Google Scholar 

  25. Ding, L. & Morrison, S. J. Haematopoietic stem cells and early lymphoid progenitors occupy distinct bone marrow niches. Nature 495, 231–235 (2013)

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  26. Greenbaum, A. et al. CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature 495, 227–230 (2013)

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  27. Zou, Y. R., Kottmann, A. H., Kuroda, M., Taniuchi, I. & Littman, D. R. Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development. Nature 393, 595–599 (1998)

    ADS  CAS  PubMed  Article  Google Scholar 

  28. Cerdan, C. & Bhatia, M. Novel roles for Notch, Wnt and Hedgehog in hematopoesis derived from human pluripotent stem cells. Int. J. Dev. Biol. 54, 955–963 (2010)

    CAS  PubMed  Article  Google Scholar 

  29. van Eeden, F. J. et al. Mutations affecting somite formation and patterning in the zebrafish, Danio rerio. Development 123, 153–164 (1996)

    CAS  PubMed  Article  Google Scholar 

  30. Higashijima, S., Okamoto, H., Ueno, N., Hotta, Y. & Eguchi, G. High-frequency generation of transgenic zebrafish which reliably express GFP in whole muscles or the whole body by using promoters of zebrafish origin. Dev. Biol. 192, 289–299 (1997)

    CAS  PubMed  Article  Google Scholar 

  31. Ma, D., Zhang, J., Lin, H. F., Italiano, J. & Handin, R. I. The identification and characterization of zebrafish hematopoietic stem cells. Blood 118, 289–297 (2011)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  32. Lawson, N. D. & Weinstein, B. M. In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev. Biol. 248, 307–318 (2002)

    CAS  PubMed  Article  Google Scholar 

  33. Proulx, K., Lu, A. & Sumanas, S. Cranial vasculature in zebrafish forms by angioblast cluster-derived angiogenesis. Dev. Biol. 348, 34–46 (2010)

    CAS  PubMed  Article  Google Scholar 

  34. Pisharath, H., Rhee, J. M., Swanson, M. A., Leach, S. D. & Parsons, M. J. Targeted ablation of beta cells in the embryonic zebrafish pancreas using E. coli nitroreductase. Mech. Dev. 124, 218–229 (2007)

    CAS  PubMed  Article  Google Scholar 

  35. Seger, C. et al. Analysis of Pax7 expressing myogenic cells in zebrafish muscle development, injury, and models of disease. Dev. Dyn. 240, 2440–2451 (2011)

    CAS  PubMed  Article  Google Scholar 

  36. Li, Q. et al. Chemokine signaling guides axons within the retina in zebrafish. J. Neurosci. 25, 1711–1717 (2005)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  37. Kwan, K. M. et al. The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs. Dev. Dyn. 236, 3088–3099 (2007)

    CAS  PubMed  Article  Google Scholar 

  38. Johnson, J.-L. F. A. et al. Scube activity is necessary for Hedgehog signal transduction in vivo. Dev. Biol. 368, 193–202 (2012)

    CAS  PubMed  Article  Google Scholar 

  39. Distel, M., Wullimann, M. F. & Koster, R. W. Optimized Gal4 genetics for permanent gene expression mapping in zebrafish. Proc. Natl Acad. Sci. USA 106, 13365–13370 (2009)

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  40. Berger, J. & Currie, P. D. 503unc, a small and muscle-specific zebrafish promoter. Genesis 51, 443–447 (2013)

    CAS  PubMed  Article  Google Scholar 

  41. Villefranc, J. A., Amigo, J. & Lawson, N. D. Gateway compatible vectors for analysis of gene function in the zebrafish. Dev. Dyn. 236, 3077–3087 (2007)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  42. Westerfield, M. The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio) 4th edn (Univ. of Oregon Press, 2000)

    Google Scholar 

  43. Curado, S., Stainier, D. Y. & Anderson, R. M. Nitroreductase-mediated cell/tissue ablation in zebrafish: a spatially and temporally controlled ablation method with applications in developmental and regeneration studies. Nature Protocols 3, 948–954 (2008)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  44. Shimshek, D. R. et al. Codon-improved Cre recombinase (iCre) expression in the mouse. Genesis 32, 19–26 (2002)

    CAS  PubMed  Article  Google Scholar 

  45. Thisse, C. & Thisse, B. High-resolution in situ hybridization to whole-mount zebrafish embryos. Nature Protocols 3, 59–69 (2008)

    CAS  PubMed  Article  Google Scholar 

  46. Neyt, C. et al. Evolutionary origins of vertebrate appendicular muscle. Nature 408, 82–86 (2000)

    ADS  CAS  PubMed  Article  Google Scholar 

  47. Weinberg, E. S. et al. Developmental regulation of zebrafish MyoD in wild-type, no tail and spadetail embryos. Development 122, 271–280 (1996)

    CAS  PubMed  Article  Google Scholar 

  48. Thompson, M. A. et al. The cloche and spadetail genes differentially affect hematopoiesis and vasculogenesis. Dev. Biol. 197, 248–269 (1998)

    CAS  PubMed  Article  Google Scholar 

  49. Liao, E. C. et al. SCL/Tal-1 transcription factor acts downstream of cloche to specify hematopoietic and vascular progenitors in zebrafish. Genes Dev. 12, 621–626 (1998)

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  50. Kalev-Zylinska, M. L. et al. Runx1 is required for zebrafish blood and vessel development and expression of a human RUNX1-CBF2T1 transgene advances a model for studies of leukemogenesis. Development 129, 2015–2030 (2002)

    CAS  PubMed  Article  Google Scholar 

  51. Seo, H. C., Saetre, B. O., Havik, B., Ellingsen, S. & Fjose, A. The zebrafish Pax3 and Pax7 homologues are highly conserved, encode multiple isoforms and show dynamic segment-like expression in the developing brain. Mech. Dev. 70, 49–63 (1998)

    CAS  PubMed  Article  Google Scholar 

  52. Karolchik, D. et al. The UCSC Genome Browser database: 2014 update. Nucleic Acids Res. 42, D764–D770 (2014)

    CAS  PubMed  Article  Google Scholar 

  53. Lindeman, L. C., Vogt-Kielland, L. T., Alestrom, P. & Collas, P. Fish'n ChIPs: chromatin immunoprecipitation in the zebrafish embryo. Methods Mol. Biol. 567, 75–86 (2009)

    CAS  PubMed  Article  Google Scholar 

  54. Mathelier, A. et al. JASPAR 2014: an extensively expanded and updated open-access database of transcription factor binding profiles. Nucleic Acids Res. 42, D142–D147 (2014)

    CAS  PubMed  Article  Google Scholar 

  55. Whitesell, T. R. et al. An alpha-smooth muscle actin (acta2/alphasma) zebrafish transgenic line marking vascular mural cells and visceral smooth muscle cells. PLoS ONE 9, e90590 (2014)

    ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank G. Kardon and B. Hogan for reading and critique of the manuscript. We also thank C.-H. Wang, F. Ellett, V. Nikolova-Krstevski and Fishcore staff for technical assistance. This work was supported by a National Health and Medical Research Council of Australia (NHMRC) grant to P.D.C. and an Australian Research Council grant to P.D.C. and G.E.H.; G.E.H. was supported by a Cancer Institute NSW (CINSW) Career Development Fellowship, R.L.S. by the CINSW and RT Hall Foundation, P.D.N. by an Australian Postgraduate Award, G.J.L. by a NHMRC Senior Research Fellowship and P.D.C. by a NHMRC Principal Research Fellowship. The Australian Regenerative Medicine Institute is supported by funds from the State Government of Victoria and the Australian Federal Government.

Author information

Authors and Affiliations

Authors

Contributions

P.D.C. (fate mapping, mutant analyses), P.D.N. (fate mapping, micro injections, mutant analyses, transgenic construct generation) and G.E.H. (mutant cloning, construct generation, in situ hybridization, micro injections, mutant analyses and cell transfections) designed and performed experiments; C.S. (in situ hybridizations, microinjection), L.M. (confocal analyses), T.E.H. (generated transgenic constructs), S.B. (histology), K.J.F. (in situ hybridization, cell transfections), D.B.G. (generated transgenic constructs) and S.A. (ChIP) performed experiments; N.J.C. and R.L.S. provided reagents; G.J.L., M.R. and J.M.P. provided reagents and assisted with revisions; P.D.C., P.D.N. and G.E.H. wrote the manuscript.

Corresponding author

Correspondence to Peter David Currie.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 cho mutants exhibit defects in appendicular muscle and secondary muscle growth.

ah, Appendicular myoblast markers are reduced in cho mutants. a, b, Expression of the migratory myoblast marker lbx1a, which is expressed in migratory fin (arrows) is significantly reduced in cho homozygotes (b) compared to wild-type siblings (a). b is a more anterior view than a, displaying the hindbrain expression of lbx1a (arrowheads), which is unaffected in cho mutants. The muscle myoblast markers myogenin (cf) and myod (g, h) are severely reduced in the fin and hypaxial muscle of cho mutants (d, f, h) compared to wild-type siblings (c, e, g). The posterior hypaxial muscle (arrowhead g, h), also exhibits an altered trajectory and reduced size in cho homozygotes. ad, 30 hpf; eh, 48 hpf. Dorsal views anterior to the left. il, Deficits in muscle formation are permanent in cho mutants, a small proportion of which are adult viable. Adult mutant fish (j) exhibit a severe reduction of muscle, which is most evident within anterior myotomes when compared to wild type (i). In cross sections stained with Azan Mallory, adult cho mutants (l) exhibit fibrotic collagen deposits (blue, arrow) and a reduction in muscle (red) when compared to wild type (k). m, n, Time-lapse analyses of wild type (m) and cho mutant (n) transgenic for Tg(α-actin:GFP). This analysis reveals that fin muscle (arrowheads) fails to form correctly in cho mutants and that the posterior hypaxial muscle (PHM) (arrows) fails to extend and is much reduced in size. Furthermore, differentiating muscle cells join the PHM from anterior somites (denoted *), which are not normally fated to do so. Oblique lateral views anterior to the left. Time (t) is in minutes. Scale bars, 50 μm. o, Quantitation of fibre number reduction in cho mutants through embryonic and early larval growth. Data are mean ± s.e.m.; significance (***P < 0.0001, **P < 0.005) is from an unpaired t-test.

Extended Data Figure 2 The cho phenotype results from a mutation in the homeobox gene meox1, which is expressed in somites.

a, The cho mutation was mapped to linkage group 12, recombinants are per 2422 meioses. A stop codon (arrow) within the meox1 ORF was identified at amino acid position 87. b, c, Antisense morpholinos against meox1 resulted in a phenocopy of the cho phenotype, including inducing the characteristic melanocyte “choker” (n = 40/45, c, arrowhead) that is absent from uninjected siblings (b). dg, An antibody to Meox1 recapitulates the mRNA expression of 10 and 26 somite embryos in genotyped wild-type (meox1+/+) siblings (d, f) and is restricted to nuclei of the anterior most cells of rostral somites in 10-somite embryos (bracket d) and the ECL in 26-somite embryos (f). Meox1 protein is undetectable in similarly staged genotyped meox1−/− homozygotes (n = 18, e, g). Blue, DAPI nuclear counterstain, scale bars 50 μm. h, Three different meox orthologues exist in the zebrafish genome. A phylogenetic tree was assembled using the Geneious program with UPGMA tree build method and bootstrap resampling using meox sequences in the human (H), bovine (B), mouse (M), rat (R), chicken (C) and zebrafish (Z) genomes. The gene mutated in choker mutants (red box) is clearly aligned as the solitary zebrafish meox1 orthologue. Two zebrafish meox2 orthologues are identified by this analysis. it, Expression analysis of meox genes. in, meox1 expression initiates in early somitogenesis (i, 5 somites) is confined to anterior presomitic mesoderm (PSM), the entirety of newly formed somites and is rapidly confined to anterior of mature somites (j, 10 somites). k, At 30 hpf expression in the tail is still evident within the entire ECL, and also with VACs ventral to the notochord (m, bracket). l, Expression is also evident in a large cluster of cells (bracket) immediately adjacent to somite 1 (arrow) at 30 hpf and in the pectoral fin (PF, arrows) at 48 hpf (n). meox2a is expressed at the end of somitogenesis in formed myotomes (o) as is meox2b (p). At 48 hpf, both meox2a (q) and meox2b (r) are expressed in fin myoblasts (arrows) and meox2b is expressed in the posterior hypaxial muscle (arrowhead, r). s, At 72 hpf, meox2a is expressed in ventral extraocular muscles (arrow). t, At 72 hpf, meox2b is expressed in specific muscles of the head (arrow). Accession numbers used: NM_001045124 Hmeox1, NM_004527 variant 1 Cmeox2, NM_001005427 Cmeox1 NM_204765, Zmeox1 NM_001002450, Zmeox2a XM_679832, Zmeox2b NM_001045124, Hmeox2 NP_005915, Bmeox2 NP_001091514, Mmeox2 NP_032610, Rmeox2 NP_058845, Bmeox1 NP_001030453, Mmeox1 NP_034921, Rmeox1 NP_001102307.

Extended Data Figure 3 The endotome contributes to the endothelial cells of the embryonic vasculature.

af, In 20 somite WT embryos (ac), cxcl12b is expressed within cells of the ventral aspect of the somites (arrows, c). In meox1−/− mutants, the ventral, somitic expression of cxcl12b is expanded though out the mediolateral, anterior/posterior and dorsoventral extent of the ventral somite (compare brackets b′ to e′, ventral views of yolk extension somites and arrows in c, f; asterisk, expression in the pronephros). g, h, At 48 hpf, cxcl12b expression in WT (g) embryos diminishes within the embryonic vasculature, while it remains high in meox1−/− mutants (h). i, j, notch3 expression, which marks the dorsal aorta (arrows), is moderately expanded in meox1 mutants (j) compared to wild type siblings (i). a, d, g, h, i, j, Lateral view, anterior to the left, c, f, cross-section dorsal to top at the level of dotted lines in a and d respectively, b, b′, e, e′ ventral view anterior to left. kk′′, Segmentation reveals computationally annotated co-localization of Meox1 protein and vascular GFP in yellow. Notably, in the dorsal aspect of the DA, co-localization occurs in DA cells immediately adjacent to the base of ISVs (arrows k′k′′) and in the initially formed cells of the vessels themselves (arrows kk′′) as well as in cells positioned ventrally in the DA (arrowheads, k′k′′). ll′′′, Single confocal slices in sagittal (l) and transverse (l′l′′′) views of the same Meox1 positive nuclei co-localizing with fli1a–GFP positive endothelial cells in the DA (yellow), blue is DAPI, green is vasculature, red is Meox1 protein. Scale bars, 50 μm. mo, Zebrafish triply transgenic for pax3a:KalTA4, uas:NTRmCherry and fli1a:GFP reveals that cells of the endotome, (marked by mCherry expression, m, m′, o, o′) contribute endothelial cells (marked by fli1a:GFP expression, m, m′, n, n′, arrows) to the embryonic vasculature including the dorsal aorta and PCV (n = 3). m′, n′, o′, Cross sections at the level of the dotted line in m, n, o, respectively. pu, Somite-specific photo conversion reveals a contribution of the endotome to the endothelial cells of the nascent vasculature. pr, Zebrafish carrying transgene for the photoconvertible nlsKaede protein expressed from the somite-specific mesogenin promoter (Tg(msgn1:nlsKaede)) that has undergone regional photoconversion in the anterior somite at the 10 somite stage encompassing the endotome (n = 11). su, Same embryo as in pr in which red represents photoconverted somite nuclei and green represents fli1a–GFP cytoplasmic expression in the embryonic vasculature. s, The total anterior somitic contribution, including muscle nuclei, is shown in red and endothelial cells of the vasculature in green. t, Same image as in s, but with non-vasculature photoconverted nuclei computationally removed, showing the nuclei of the endotome contributing to the nascent vasculature in yellow. u, Computationally generated somite-derived nuclei that contribute to the vasculature shown alone in red. vx′, Zebrafish triply transgenic for pax3a:KalTA4, uas:NTRmCherry and itga2b:GFP reveal a close association between somite-derived endothelial cells marked in red, and HSCs in green but no colocalization (arrow) (n = 4). v′, w′, x′ are transverse sections in v, w and x, respectively. w, Higher magnification view of the region boxed in w. Brackets highlight DA, lines reveal transverse section level. Scale bars, 50 μm.

Extended Data Figure 4 Fidelity of somite-specific photoconversion.

a, Views in 3 dimensions (t, transverse; co, coronial; s, sagittal) of a 10-somite stage embryo into which mRNA encoding Kaede protein has been injected and photoconverted reveals somite-specific photoconversion. b, A similar analysis of the Tg(msgn1: nlsKaede) line reveals the somite-specific nature of the photoconversion. Dotted lines mark the neural tube, solid circle is notochord. ce, Maximum projection of a photoconversion of a 15-somite stage embryo carrying both the msgn1:nlsKaede and the fli1a:GFP transgenes. This shows that there is no photoconversion (red) of any cells in the nascent vasculature (marked by the solid line of cytoplasmic GFP ventral to the somites, arrows) and that photoconversion is somite-specific in this line. fh, Views in 3 dimensions (t, transverse, co, coronial, s sagittal) of the same embryo in ce showing the somite-specific nature of the photoconversion and the lack of any photoconverted cells in the endothelial cells of the nascent vasculature (arrow). Scale bars, 50 μm.

Extended Data Figure 5 The posterior cardinal vein is transiently expanded in meox1 mutants.

ad, Haematoxylin and eosin stained sections of wild-type (WT, n = 3) sibling (a, c) and meox1 mutant embryos (n = 5, b, d) at 30 hpf show an increase in size of the posterior cardinal vein (PCV, boxed, dotted circle). c, Higher magnification of region boxed in a. d, higher magnification of the region boxed in b. ef. In situ hybridization with foxc1a at 30 hpf, which marks the perivascular space, highlights the increase in cross sectional area of the PCV in meox1 mutants (n = 10, f) compared to WT siblings (n = 30, e). g, h, Lateral views of fli1a–GFP transgenic zebrafish counterstained with DAPI (blue) reveals an expansion of the PCV (brackets) in meox1 mutants (n = 7, h) compared to WT (n = 9, g). i, j, Cross sections at the level of the yolk extension of fli1a–GFP transgenic zebrafish stained with an antibody against Myosin Heavy Chain (MyHC, red), where the reduction in secondary muscle is also apparent in meox1 mutants as well as the expansion in the PCV, brackets show diameter of PCV. ko, The expansion in the size of the PCV is transient. Transgenic zebrafish carrying both the fli1a–GFP and the kdrl–mCherry transgenes, which allows the PCV versus aorta to be more clearly distinguished due to differential level of expression of these transgenes, reveals that at 3 days post fertilisation (dpf) (k, l) the expansion in size of the PCV remains evident in meox1 mutants (l) compared to WT (k) but is not detectable in meox1 mutants (o) compared to WT (m) at 6dpf. p, Quantitation of the diameter size of the dorsal aorta and PCV at 3 and 6 dpf reveals a statistically significant increase in the size of the PCV at 3dpf but not at 6 dpf and no change in dorsal aorta size over a similar time. (n = 6 for each time point and genotype, insets ko, cross sections at the level of the dotted line, brackets mark the PCV) (**P < 0.005). Scale bars, 50 μm.

Extended Data Figure 6 meox1 mutant embryos do not exhibit an increase in the number of cells expressing phosphohistoneH3 (PHH3).

ae, An examination of PHH3 positive nuclei at 15 somites (a, b) and 22 somites (c, d) in wild type (a, c) and meox1 mutants (b, d) reveals no increase in proliferation within the somites of meox1 mutants (e). There was in fact a significant drop in somite proliferation in meox1 mutants at 15 somites, which may reflect the exit of the excess endotomal cells from the somite at this stage. Data are mean ± s.e.m.; significance (*P < 0.05) is from an unpaired t-test. Blue, DAPI nuclear counterstain; scale bars, 20 μm. We also retrospectively examined the number of cells that generated DA cells and VACs in meox1 mutant and wild-type embryos when labelled iontophoretically. This analysis also showed there was no significant difference in the number of cells generated from individual single somitic cell labels in either wild-type (1 somite cell generates 1.167 ± 0.1 s.e.m. clonal progeny n = 38) or meox1 mutant/morphant embryos (1 somite cell generates 1.20 ± 0.93 s.e.m. clonal progeny n = 34) despite their different sized ‘vascular competent’ endotome.

Extended Data Figure 7 Sustained meox1 activity promotes muscle progenitor formation and suppresses endothelial fate.

al, Injection of meox1 mRNA. ah, meox1 mRNA injection induces expression of pax7a. At 10 somites, pax7a transcript localization is expanded from its wild-type expression in anterior border cells (a) to the entire AP extent of the somite (b) (38% n = 26). c, High magnification view of the region boxed in a. d, High magnification view of the region boxed in b. e, f, Lateral views, anterior to the left, of 15-somite stage embryos injected with meox1 mRNA show a dorsal-ventral expansion of pax7a expression (bracket, f) when compared to uninjected sibling (bracket, e). g, h, meox1 injection induces ectopic pax7a expression in 26 somite injected embryos (h) when compared to wild type (g) (26%, of injected embryos n = 31). il, Injection of meox1 mRNA into the Tg(fli1a–GFP) line specifically inhibits ISV sprouting (n = 37, j, l) when compared to uninjected siblings (i, k). k is a high magnification view of the region boxed in i. l is a high magnification view of the region boxed in j. mv′, Ectopic expression of meox1 in the embryonic vasculature can cell autonomously induce Pax7 in a subset of cells and disrupt vasculature formation. mq′, Expression of mCherry (red) from the fli1a promoter does not induce vascular malformations when injected in the fli1a–GFP line (m, n) and does not induce ectopic Pax7 (Green) oq′. rv′, By contrast expression of Meox1–mCherry fusion (red) but not mCherry alone from the fli1a promoter induced vascular malformations, specifically inhibiting ISV formation when injected in the Tg(fli1a–GFP) line (rs) and induces ectopic Pax7 within vascular cells (arrows, n = 15, 18%). tv′, Blue, DAPI nuclear counterstain, scale bars 50 μm.

Extended Data Figure 8 Meox1 deficient embryos possess a marked increase in haematopoietic stem cells.

ai′, Injection of a meox1 morpholino results in an increase in itga2b–GFP positive cells at 30 hpf (bc′) and 48 hpf, both in Dorsal Aorta (DA) and the Caudal Haematopoietic Tissue (CHT) (ef′, hi′), compared to uninjected embryos (a, a′ 30 hpf, d, d′, g, g′ 48 hpf). Scale bars 50 μm. jo, The number of runx1 positive cells (arrows, k, l, n, bracket o) is expanded in meox1 mutants (n = 26, mo) compared to wild-type siblings in 48 hpf embryos (jl). k and l, high magnification views of the areas boxed in j. n and o, high magnification views of the areas boxed in m. pu, At 30 hpf, meox1 deficient embryos (n = 32, su) exhibit an expanded population of DA associated HSC marker cmyb when compared to WT siblings (pr). q, r, higher magnification of regions boxed in p. t and u higher magnification of regions boxed in s. v, Quantification of itga2b–GFP low cells. 30 hpf are counts for the region of the DA underlying the first 6 somites post yolk extension. 48 hpf DA are itga2b–GFP positive cell counts for the region of the DA underlying the first 6 somites anterior to the end of the yolk extension. 48 hpf CHT, Increased colonization of HSCs to the CHT is also observed, in line with the expected consequence of increased numbers of DA-specified HSCs evident in meox1 mutants. CHT counts are for the region underlying the first 6 somites posterior to the yolk extension. w, meox1 morpholino injection results in a significant increase in the percentage of phospho histone H3 positive (PHH3+) itga2b–GFP positive HSCs present in the DA at 36 hpf. Data are mean ± s.e.m.; significance (***P < 0.0001, **P < 0.005, *P < 0.05) are from an unpaired t-test.

Extended Data Figure 9 Somite-derived endothelial cells do not contribute to HSCs in wild type and meox1 mutant embryos.

af, Tg(msgn1:nlsKaede) crossed to Tg(itga2b:EGFP). Photoconversion of nuclear Kaede expressed exclusively within the somites using a mesogenin promoter at the 10 somite stage (a) generates muscle and vascular associated nuclei at 26 somites (b). c, d, Maximum projections of the vascular region of the same embryo shown in b revealing an association but a lack of co-localization between somite-derived ECs and itga2b–EGFP, c is lateral view, d is transverse view. e, f, Single confocal slice with DIC (e) or without (f), revealing that somite-derived EC are intimately associated with itga2b–EGFP positive HSCs but do not co-localize with them (n = 40). gn′, Iontophoretic fate mapping in wild-type (n = 23, gj) and meox1 morpholino injected embryos (n = 35, kn′) reveals that under time-lapse individually labelled cells derived from the endotome are intimately associated with multiple itga2b–EGFP expressing cells, but never themselves express itga2b–EGFP. g, k, Dorsal view 10 somite stage, tetramethylrhodamine dextran label (red) within the endotome in wild-type (g) and meox1 morpholino injected embryos (k). Same embryos at 36 hpf (h, l), 60 hpf (i, j, m) and 84 hpf (n) lateral view anterior to the left. g, h, km, Live images. i, j, n, Single confocal scans. i, j, n, Maximum projections; i, n, lateral view; j, cross-sectional view, arrows mark labelled ECs. Lines mark vasculature (i) and circle represent the position of the DA (j). n′, high magnification view of the region boxed in n. ox, wnt16 expression is not expanded in meox1 mutants. wnt16 expression becomes restricted to anterior dorsal anterior sector (arrows p, q) of the zebrafish somite after an initially broad somitic expression at the 15-somite stage (bracket r), but is undetectable in meox1 mutants (s, t). o, p, rt, Dorsal views anterior to the left. q, Lateral view, anterior to the left. p, r and t, High magnification views of the regions boxed in o and s. Dotted lines outline somite boundaries. ux, wnt16 expression at the 20 somite stage is restricted to a dorsal anterior section of the external cell layer (ECL) in WT embryos. u′ cross section at the level of the line marked in u, ECL arrows. v, High magnification view of the region marked in u. w, In meox1 mutants, wnt16 is severely reduced and detected in only a very few cells ectopically located at the posterior medial section of individual somites (arrows, x). x, High magnification view of the region boxed in w.

Extended Data Figure 10 Meox1 acts as a repressor and directly regulates the expression of cxcl12b in zebrafish.

ad′, Reduction in cxcl12 signalling rescues the meox1−/− phenotype. Addition of AMD3100 (20 μM b, b′) or injection of with a morpholino against the cxcl12b gene (c, c′) prevents the expansion of HSCs evident in meox1 mutants (a, a′). Expression of cxcl12b from the endotome specific pax3a promoter induces clustering of HSC. d, d′, Clustering of cmyb positive HSCs is induced upon expression of a cxcl12b protein fused to mCherry from the endotome specific pax3a promoter. e, cxcl12b expressed on endotome-derived endothelial cells rescues the HSC deficit in embryos lacking global Cxcl12b activity. unject, uninjected siblings embryos. morp: Embryos injected with a cxcl12b morpholino, designed against the 5′UTR of cxcl12b gene. This morpholino knocks down global endogenous Cxcl12b activity in the developing embryo in a similar manner to the original ATG targeted morpholino, but does not target the open reading frame of cxc12b present within the pax3a:cxcl12b mCherry construct. morp. pax3a–mCherry: Embryos injected with the 5′UTR morpholino and pax3a mCherry vector only control that does not contain the cxcl12b open reading frame. morp. pax3a;cxcl12bmcherry: Embryos injected with the 5′UTR morpholino and a construct expressing cxcl12b fused to mCherry from the pax3a promoter. In these injected animals, endotome-derived endothelial cells expressing cxcl12b are able to rescue the deficit of global knock down of cxcl12b. Counts are total cmyb positive (cmyb+ve) HSCs in the DA, six somites anterior to the end of the yolk extension. Data are mean+ s.e.m., significance **P < 0.005 from an unpaired t-test. f, Human aortic endothelial cells (HAECs) were transfected with an expression construct in which Meox1 is fused to GFP. Cells were cultured for 24 h and sorted by FACs for GFP expression. Real time PCR was used to determine the relative level of expression of cxcl12 to ribosomal protein, large, P0 (RPLP0). Fold expression changes between GFP-positive and GFP-negative cells were calculated from three separate transfection experiments. Data are mean ± s.e.m. transfections n = 3, PCR on each transfection performed in triplicate. g, UCSC genome browser showing alignment using the Phastcons algorithim of cxcl12b across 5 fish species revealing specific conserved non-coding elements (CNEs). All alignment algorithms tested (Phastcons, Phylof, Multiz) reveal the same seven (labelled 1–7) CNEs within introns and approximately 6.5 kb upstream of the ATG start site. Possum software was consequently used to locate Meox1 binding sites within these regions using the Meox1 positional weight matrix (PH0103.1) obtained from the Jaspar database54. Of the 7 selected regions only 4 (red boxes) had matching Meox1 binding sites. h, Chromatin immunoprecipitation (ChIP)-qPCR analysis of the 4 selected regions was conducted with 3 non-overlapping primer sequences spanning the conserved region, and showed Meox1 occupancy is enriched in regions 1, 2 and 6 but is not found at region 7. Data are mean+ s.e.m. i, A model for endotome formation and its contribution to definitive HSC induction. A number of distinct phases contribute to the induction of haematopoietic stem cells by somite-derived endothelial cells. (i), The somite is initially separated into a primary posterior myogenic domain (red) and a non myogenic anterior compartment (pink). (ii), The anterior somitic compartment is further partitioned into either endotome (orange) or dermomyotome (yellow) by the activity of the meox1 gene. Endotome cells expressing the chemokine cxcl12b migrate from the somite. (iii), The dorsal aorta (DA) is colonized by endotome cells, which contribute to endothelial cells and a second set of cells termed vascular associated cells (VACs). A potential fate of VACs could be mural cells, which give rise to vascular associated pericytes and smooth muscle cells of the vasculature. This is unlikely, however, as DA associated expression of smooth muscle actin, which specifically marks both these cell types, does not occur until 3dpf, a phase of development well after the processes we observe here55. (iv), HSC (green) induction requires cxcl12b activity (arrowheads) secreted from endotome-derived endothelial cells. NT, neural tube; NC, notochord; DA, dorsal aorta; PCV, posterior cardinal vein.

Supplementary information

Migration of fin and posterior hypaxial muscle in wildtype sibling embryos

12 hr timelapse of alpha actin-GFP transgenic wildtype sibling embryos from 30hpf of development. (AVI 2093 kb)

Migration of the fin and posterior hypaxial muscle in choker homozygous mutant embryo

12 hr timelapse of alpha actin-GFP transgenic cho mutant embryos from 30hpf of development. (MOV 363 kb)

Photoconversion reveals a somitic contribution to the vasculature in zebrafish

Video of the initial photoconversion at the 10 somite stage of nuclear Kaede. Rotation of photoconversion followed by surface rendering of maximum projection to highlight the location of photoconverted nuclei. (AVI 10127 kb)

Photoconversion reveals a somitic contribution to the vasculature in zebrafish

Video of the same embryo in Video 3 revealing the somitic contribution to the vasculature marked with cytoplasmic GFP in the fli1a-GFP transgenic line. Rotation of maximum projection for photoconverted cells followed by surface rendering of the same image to highlight vasculature (green) and co-localised nuclear Kaede within the vasculature (yellow). (AVI 5348 kb)

Photoconversion of a mesogenin nuclear Kaede transgenic line

Video of the photoconversion of the anterior somite at the 10 somite stage of the somite specific msgn-nlsKaede line illustrates the somite specificity of expression and ease of photoconversion evident in this line. (AVI 2611 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Nguyen, P., Hollway, G., Sonntag, C. et al. Haematopoietic stem cell induction by somite-derived endothelial cells controlled by meox1. Nature 512, 314–318 (2014). https://doi.org/10.1038/nature13678

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nature13678

Further reading

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

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