Skip to main content

Thank you for visiting 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.

Auto-attraction of neural precursors and their neuronal progeny impairs neuronal migration


Limited neuronal migration into host brain tissue is a key challenge in neural transplantation. We found that one important mechanism underlying this phenomenon is an intrinsic chemotactic interaction between the grafted neural precursor cells (NPCs) and their neuronal progeny. NPCs secrete the receptor tyrosine kinase ligands FGF2 and VEGF, which act as chemoattractants for neurons. Interference with these signaling pathways resulted in enhanced migration of human neurons from neural clusters.

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

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Get just this article for as long as you need it


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

Figure 1: Purified neurons exhibit a vastly enhanced migration potential.
Figure 2: Chemotactic interaction between neural precursor cells and their neuronal progeny.
Figure 3: Inhibition of NPC-neuron auto-attraction results in enhanced migration in CNS tissue.


  1. Koch, P., Kokaia, Z., Lindvall, O. & Brustle, O. Lancet Neurol. 8, 819–829 (2009).

    Article  Google Scholar 

  2. Lindvall, O. & Kokaia, Z. Trends Pharmacol. Sci. 30, 260–267 (2009).

    Article  CAS  Google Scholar 

  3. Barker, R.A., Barrett, J., Mason, S.L. & Bjorklund, A. Lancet Neurol. 12, 84–91 (2013).

    Article  CAS  Google Scholar 

  4. Fricker, R.A. et al. J. Neurosci. 19, 5990–6005 (1999).

    Article  CAS  Google Scholar 

  5. Roy, N.S. et al. Nat. Med. 12, 1259–1268 (2006).

    Article  CAS  Google Scholar 

  6. Tabar, V. et al. Nat. Biotechnol. 23, 601–606 (2005).

    Article  CAS  Google Scholar 

  7. Svendsen, C.N. & Caldwell, M.A. Prog. Brain Res. 127, 13–34 (2000).

    Article  CAS  Google Scholar 

  8. Kinouchi, R. et al. Nat. Neurosci. 6, 863–868 (2003).

    Article  CAS  Google Scholar 

  9. Reier, P.J., Perlow, M.J. & Guth, L. Brain Res. 312, 201–219 (1983).

    Article  CAS  Google Scholar 

  10. Koch, P., Opitz, T., Steinbeck, J.A., Ladewig, J. & Brustle, O. Proc. Natl. Acad. Sci. USA 106, 3225–3230 (2009).

    Article  CAS  Google Scholar 

  11. Ladewig, J. et al. Stem Cells 26, 1705–1712 (2008).

    Article  Google Scholar 

  12. Zicha, D., Dunn, G.A. & Brown, A.F. J. Cell Sci. 99, 769–775 (1991).

    PubMed  Google Scholar 

  13. Hilberg, F. et al. Cancer Res. 68, 4774–4782 (2008).

    Article  CAS  Google Scholar 

  14. Shi, Y., Kirwan, P., Smith, J., Robinson, H.P. & Livesey, F.J. Nat. Neurosci. 15, 477–486 (2012).

    Article  CAS  Google Scholar 

  15. Tailor, J. et al. J. Neurosci. 33, 12407–12422 (2013).

    Article  CAS  Google Scholar 

  16. Richards, K.L. & McCullough, J. Immunol. Commun. 13, 49–62 (1984).

    Article  CAS  Google Scholar 

  17. Foxman, E.F., Kunkel, E.J. & Butcher, E.C. J. Cell Biol. 147, 577–588 (1999).

    Article  CAS  Google Scholar 

  18. Opitz, T., Scheffler, B., Steinfarz, B., Schmandt, T. & Brustle, O. Nat. Protoc. 2, 1603–1613 (2007).

    Article  CAS  Google Scholar 

  19. Scheffler, B. et al. Development 130, 5533–5541 (2003).

    Article  CAS  Google Scholar 

  20. Cao, X. et al. Immunity 2, 223–238 (1995).

    Article  CAS  Google Scholar 

Download references


We thank J. Itskovitz-Eldor (Technion, Israel Institute of Technology) for providing the human embryonic stem cells (line H9.2) used for generating the neural precursor cells, B. Steinfarz for preparing and culturing the hippocampal slice cultures, A. Leinhaas for support with the in vivo experiments, E. Endl at the flow cytometry facility of the University Hospital Bonn for cell sorting, the National Institute of Allergy and Infectious Diseases for contributing the C57BL/10SgSnAi[KO]γc-[KO]Rag2 mice, and J. Tailor and A. Smith (Wellcome Trust-Medical Research Council Stem Cell Institute) for providing human fetal neural stem cells. This work was supported by the European Union (grant 222943 Neurostemcell; LSHG-CT-2006-018739, ESTOOLS), the German Research Foundation (DFG; SFB-TR3) and the Hertie Foundation.

Author information

Authors and Affiliations



J.L. and P.K. conceived and designed the study, performed the experiments, assembled, analyzed and interpreted the data, and wrote manuscript. O.B. conceived and designed the study, assembled, analyzed and interpreted the data, and wrote the manuscript.

Corresponding author

Correspondence to Oliver Brüstle.

Ethics declarations

Competing interests

O.B. is a co-founder of and has stock in LIFE and BRAIN GmbH.

Integrated supplementary information

Supplementary Figure 1 Experimental design and in vitro characterization of GFP-positive and GFP-negative FACSorted populations.

(a) Experimental design: a doublecortin (DCX)-GFP NPC reporter line was differentiated by growth factor withdrawal. The pre-differentiated population consisting of approximately 30% DCX-GFP-positive neurons and 70% less differentiated NPCs (30% neurons / 70% NPCs; NPC-diff) was either taken directly for migration assays or subjected to FACSorting to generate purified immature DCX-GFP-positive neurons (purified neurons). Migration of neurons within the NPC-diff population was compared to the migration of purified neurons. (b) FACSorted cells co-express GFP and DCX. (c) 24 h after FACSorting and replating, GFP-positive sorted cells maintain strong GFP fluorescence and are largely negative for nestin. (d) GFP-positive cells co-express the neuronal marker ßIII tubulin. (e) FACS sorting does not affect neuronal identities. Similar numbers of the neurons within the NPC-diff population (black bars) and the FACS-sorted population (white bars) express the transcription factors Lim1/2 and Lim3 (both markers are preferentially expressed by GABAergic interneurons of the hindbrain, the preferential fate of our GFP NPC reporter line10. (f-k) GFP-negative cells express markers of early neural precursor cells and not terminal differentiated glial cells. Analysis of the GFP negative fraction revealed homogenous expression of the neural precursor markers Nestin (f) and Sox2 (g). Many of the replated NPCs also co-express the rosette-associated markers Dach1 (h) and PLZF (i). (j-k) RT-PCR analyses revealed expression of neural stem and precursor markers (j) while glial markers could not be detected (k). RNA from human embryonic stem cell-derived neural rosettes (Rosette) or human fetal brain (FB) served as control. Bands were cut from full-length gels. Scale bars b, d, f 50 μm, c 100 μm, g-I 25 μm.

Supplementary Figure 2 Emigration of neurons from neural grafts depends on NPC content

(a) Purified GFP-positive neurons were mixed at different ratios with NPCs not carrying the DCX-GFP reporter and deposited on hippocampal rat slice cultures (experimental design: upper panel). Seven days after deposition, slice cultures were stained for Nestin and GFP. The spread of GFP-positive neurons correlates negatively with the percentage of NPCs in the cell population. (b-g) Representative pictures showing transplants containing 30% NPCs (b-d) and 5% NPCs (e-g). Note that GFP-positive neurons show a wider distribution when 5% NPCs are present in the cell preparation as compared to cell preparations containing 30% NPCs both, on top of the slice (b-c, e-f) as well as in confocal reconstructions illustrating migration of neurons into the slice tissue along the z-axis (d, g). (h) Quantification of GFP-positive neurons found in the horizontal plane outside a 250 μm perimeter around deposits containing admixtures of +30% or +5% NPCs. Shown are relative numbers of neurons in relation to the +30% NPCs admixture (set to 1; unpaired t-test, *P < 0.01). (I) Quantification of GFP-positive neurons in the vertical plane. Shown are relative numbers of neurons reaching a mid- or bottom level location from deposits containing purified GFP+ neurons plus 5% NPCs (grey) or purified GFP+ neurons only (white) in relation to deposits composed of purified GFP+ neurons plus 30% NPCs (black, set to 1; unpaired t-test, *P < 0.01). Data are based on n = 3 independent experiments per condition and shown as mean + SD Scale bars: a 1000 μm b,e 850 μm c,f 100 μm.

Supplementary Figure 3 Auto-chemoattraction between neural precursor cells and immature neurons

(a) Quantitative RT-PCR analysis of FGF2 and VEGF expression in proliferating DCX-GFP NPCs and in the FACSorted GFP-negative fraction 6, 12 and 18 days after growth factor withdrawal reveals stable expression of both chemoattractants. (b) RT-PCR analysis of FGF and VEGF receptor expression in purified neurons. Human fetal brain tissue (FB) served as control. Bands were cut from full-length gels. (c-i) Scatter diagrams depicting migration tracks in a Dunn chamber assay after a 6 h time period. The starting point of each cell is the intersection of the X and Y axes (0.0). The data points indicate the final position of individual cells at the end of the recording period. (c) Control experiment showing migration in medium only. For assessing chemokinesis equal concentrations of (d) FGF2 or (e) VEGF were added to the inner and the outer well of the chamber. Chemotaxis was assessed by placing (f) FGF2 or (g) VEGF in the outer well only. The Y-axis depicts the gradient towards the growth factors. (h-i) Chemotaxis / directionality of migration towards FGF2 and VEGF is lost upon treatment with BIBF1120.

Supplementary Figure 4 BIBF1120 has no toxic effect on NPCs or neurons, and does not influence the differentiation potential of NPCs and the phenotype of the neurons derived thereof

(a) Quantification of cytotoxicity and cell viability upon treatment of purified neurons and NPCs with BIBF1120 (both in relation to untreated controls). Data are based on the CytoTox-ONETM homogeneous membrane integrity assay and the CellTiter-Glo® luminescent cell viability assay. (b-f) Effect of BIBF1120 on neuronal differentiation of NPCs. (b) Percentage of beta-III tubulin-positive neurons in NPCs cultured for 10 days in neuronal differentiation medium with and without BIBF1120. (c) Percentage of Lim1/2 and Lim3 expressing GFP+ neurons in DCX-GFP lt-NES cells cultured for 4 weeks in neuronal differentiation medium with and without BIBF1120. Co-staining of GFP with (d) Lim1/2 or (e) Lim3 four weeks following differentiation of DCX-GFP NPCs in the presence of BIBF1120. (f) We previously described that neurons derived from the DCX-GFP NPC population are of a mainly GABAergic neurotransmitter phenotype10,11. This neurotransmitter phenotype is maintained following BIBF1120 treatment. Scale bars e-f 20 μm, d 200 μm.

Supplementary Figure 5 Characterization of hESC-derived telencephalic and human fetal brain derived neural precursor cells

(a) HESC-derived anterior NPCs show homogenous expression of the NPC-markers Pax6 and Sox2. (b) The majority of the NPCs also express the rosette-associated marker PLZF and show typical rosette-like architectures with pronounced expression of ZO1 in the center of the rosettes. (c-d) Homogenous expression of the forebrain-associated markers Otx2 (c) and FoxG1 (d) confirms the anterior identity of the NPCs. (e) Upon differentiation the NPCs give rise to a dominant fraction of neurons expressing ß-III tubulin and DCX. (f) More mature NPC-derived neurons express vGlut1, a marker for glutamatergic neurons. (g) A subfraction of the neurons express Ctip2, a transcription factor expressed in cortical layer V neurons. (h) GABAergic neurons could be detected only occasionally (white arrow). (i-j) Human fetal brain derived NPCs show homogenous expression of the NPC-markers Sox2, Pax6 and Nestin. (k) The majority of the NPCs also express the rosette-associated marker PLZF and show typical rosette-like architectures with pronounced expression of ZO1 in the center of the rosettes. (l) Upon differentiation the NPCs give rise to a dominant fraction of neurons expressing ß-III tubulin and DCX. Scale bars: a-e, i-j, l 100 μm, f-h, k 50 μm.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–5 and Supplementary Tables 1,2 (PDF 6566 kb)

Supplementary Table 1

Primers for analysis of NPCs, neuronal chemoattractants and their receptors (DOC 53 kb)

Supplementary Table 2

Antibodies used in this study (DOC 35 kb)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Ladewig, J., Koch, P. & Brüstle, O. Auto-attraction of neural precursors and their neuronal progeny impairs neuronal migration. Nat Neurosci 17, 24–26 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


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