Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 1.
Published in final edited form as: Methods Mol Biol. 2019;2002:1–11. doi: 10.1007/7651_2018_180

In vitro maintenance of multipotent neural crest stem cells as crestospheres

Sofie Mohlin 1, Laura Kerosuo 2,3,*
PMCID: PMC8014252  NIHMSID: NIHMS1065344  PMID: 30159826

Abstract

Neural crest cells are a critical source of many cell types of the vertebrate body. However, as a stem cell population they are peculiar because of the transient nature of their stem cell niche; soon after the multipotent neural crest cells are specified in the neuroepithelium, they become mesenchymal cells that migrate into various destinations in early embryos. These rapid in vivo changes during neural crest development complicate studies of their stem cell properties. Crestospheres are in vitro maintained primary cultures of premigratory neural crest cells that maintain a mixture of neural crest stem- and progenitor cells for weeks without spontaneous differentiation, including the multipotent neural crest stem cells. Here we describe how crestosphere cultures are initiated from either cranial or trunk levels of chick embryos. Alternatively, the same culture conditions can be used to maintain human embryonic stem cell derived neural crest cells as crestospheres. Thus, crestospheres provide a useful tool for studies of neural crest stemness.

Keywords: Chick embryo, crestospheres, neural crest, cranial neural crest, trunk neural crest, multipotency, self-renewal, human ES-cell derived neural crest

1. Introduction

The neural crest is a transient stem cell population that arises from the developing neural ectoderm at all axial levels in vertebrates. Neural crest cells are multipotent, giving rise to melanocytes as well as peripheral neurons and glia. Additionally, specific regions such as the cephalic, cardiac or trunk neural crest cells give rise to axial level specific cell types such as cranial bone and cartilage, the cardiac septum of the outflow tract of the heart, or the endocrine cells of the adrenal medulla. These different axial levels have been shown to be actively maintained by distinct cranial and trunk gene regulatory circuits [1], yet many open questions remain regarding how neural crest stemness is maintained. According to current understanding, neural crest cells are highly multipotent at the premigratory stage in the dorsal neural tube, and gradually lose their stem cell potential during the course of migration [2], but very little is known about the details of this process (Fig 1A).

Figure 1.

Figure 1.

Open in a new tab

Crestosphere technique. A) According to current understanding, neural crest cells are highly multipotent at the premigratory stage of their development immediately after specification at the dorsal neural tube. The transient niche is left empty after the neural crest cells become mesenchymal and migrate to multiple destinations in the vertebrate body and gradually lose their stemness. B) Cartoon visualizing different sources of crestosphere cultures and C) a list of applications (modified from Kerosuo et al., 2015). Reprinted from “Crestospheres: Long-Term Maintenance of Multipotent, Premigratory Neural Crest Stem Cells”, Vol. 5/ Issue 4, Kerosuo L, Nie S, Bajpai R, Bronner ME, pp. 499–507, Copyright 2015, with permission from Elsevier.

Cell culture models are useful tools to address mechanistic questions about stem cell fate. Primary neural crest cell cultures have been established for decades [3–6] but the cultures have been initiated with migrating neural crest cells under conditions that promote spontaneous differentiation, thus restricting the maintenance of self-renewal and multipotency. Crestosphere cultures “stop time” and maintain neural crest cells in a multipotent stem cell state providing a powerful tool for studies on the characteristics of premigratory neural crest stem cells [2]. They can be utilized for studying self-renewal, proliferation, differentiation or cell death assays, and their developmental potential can also be addressed by performing transplantation into donor embryos following in vitro manipulation [7,8]. Crestospheres are well suitable for various molecular biology applications including determination of gene expression, epigenetics or protein-protein interactions, and to study neural crest cell response to various stimuli induced by either modifications in the cell culture medium or by using gene perturbation techniques.

We have established protocols for both cranial- [7] and trunk-derived crestospheres [14]. In brief, crestospheres are established from dissected and dissociated neural tubes from embryos that are at the premigratory stage of their neural crest development (4–7 somite stage for cranial and 16–20 somite stage for trunk). Pools of premigratory neural crest containing neural tubes from 4–6 embryos constitute one crestosphere culture that can then be expanded and kept in culture for several weeks. Crestospheres are cultured as free-floating neuroepithelial 3D clusters of neural crest stem- and progenitor cells, similar to neurosphere cultures [9], under minimum adherence and serum free conditions with a modified protocol for optimal dorsal neural tube/neural crest cell identity and growth. Here, we describe the detailed procedure of successful establishment of crestospheres from both cranial and trunk neural crest (Fig. 1B–C).

2. Materials

Ringer’s and PBS solutions can be stored at room temperature. Basic neural crest (NC) medium should be stored at +4°C and preferably pre-warmed to 37°C before use.

2.1. Embryos

  1. Commercially purchased fertilized chicken eggs.

  2. Incubator at 37.5°C/100°F (see Note 1).

2.2. Isolation of neural tubes

  1. Seventy percent EtOH.

  2. Surgical instruments including forceps, fine scissors and micro scissors.

  3. Autoclaved (optional) Whatman filter paper squares with punched holes (use a regular paper puncher for office use) in the middle.

  4. Ringer’s solution: (Solution-1: 144g NaCl, 4.5g CaCl•2H2O, 7.4g KCl, ddH2O to 500ml; Solution-2: 4.35g Na2HPO4•7H2O, 0.4g KH2PO4, ddH2O to 500ml (adjust final pH to 7.4)) (see Note 2).

  5. Sterile PBS: (NaCl, KCl, Na2HPO4, KH2PO4, H2O).

2.3. Culture

  1. Basic NC medium: DMEM with 4.5g/L glucose, 1% penicillin/streptomycin, chick embryo extract (7.5%), B27 (1X), insulin growth factor –I (IGF-I, 20 ng/ml), basic fibroblast growth factor (bFGF, 20 ng/ml) (see Note 3).

  2. (Optional) In-house produced chick embryo extract: 11 days old chick embryos, DMEM, Gauze, 10ml syringe, hyaluronidase, ultracentrifuge (see Note 4).

  3. Additions to NC medium – 60nM (cranial) or 180 nM (trunk) retinoic acid (see Note 5) and 25 ng/ml (trunk) bone morphogenetic protein (BMP)-4.

  4. Ultra-low attachment T25 culture flasks (see Note 6).

  5. Sterile-filtered Accutase® solution for cell culture.

3. Methods

All work should be carried out in a clean environment. Dissociation of dissected neural tubes and all work thereafter should be carried out in sterile cell culture hood.

3.1. Isolation of neural tubes

  1. Incubate fertilized eggs to 4–8 somite stage (Hamburger Hamilton HH stage 8-/9 for cranial crestospheres) or 16–20 somite stage embryos (HH stage 13-/14- for trunk crestospheres).

  2. Spray eggs with 70% EtOH.

  3. Open eggs by cracking the shell from 1/3 below the top by using robust forceps. Pour out albumin through this hole and use forceps to pull out thick albumin. Gradually remove all albumin and gently wipe off remaining thick albumin surrounding the embryo using Kim Wipes tissue paper. If the embryo is on the side, gently roll the yolk by using the robust forceps (importantly avoid touching the embryo). Collect the embryo by placing a squared 1cm × 1cm filter paper on top of the embryo so that the embryo is within the stenciled non-paper area in the middle, and the vitelline membrane will then be attached to the paper. Cut off the filter paper from the yolk and transfer the embryo to Ringer’s solution in a petri dish.

  4. By careful dissection under a microscope, isolate neural tubes in Ringer’s solution from neighboring tissue using micro scissors. Start by placing the embryo ventral side up and cut the endoderm from the middle as if you were “opening a jacket”. Now carefully cut out the neural tube, and importantly exclude all neighboring mesoderm and the notochord.

  5. For cranial neural tubes, exclude the very anterior tip (that does not produce neural crest), and collect the neural tube up to second (2nd) somite level. For trunk, aim for the neural tube between the tenth and fifteenth (10–15) somite level. Each established culture should consist of 4–6 pooled neural tubes (see Note 7).

  6. Mechanically dissociate the complete pool of neural tubes by pipetting up and down 30 times in 50μl Ringer’s solution or 1XPBS using a p200 tip in an eppendorf tube. Monitor under a microscope that the tissue has been dissociated into small (20–50 cell) clumps.

  7. Prepare basic NC medium in an Ultra-low attachment T25 flask and add retinoic acid (cranial and trunk) and BMP-4 (trunk) appropriate to total volume (concentration of retinoic acid dependent on axial level source of neural tubes) (see Note 8). Transfer dissociated tissue pieces to 1–2ml prepared NC medium supplemented with retinoic acid (cranial and trunk) and BMP-4 (trunk) (see Note 9).

  8. When culturing human embryonic stem (ES) cell-derived crestospheres, induce neural crest development according to previously published protocol [10]. At day 7–9 in neural crest inducing medium, transfer the neuroepithelial spheres to NC medium in Ultra-low attachment T25 flasks.

3.2. Crestosphere cultures

  1. (Optional) Prepare in-house produced chick embryo extract by rinsing headless 11 days old chick embryos with cold DMEM on a double layer of Gauze on a 500ml beaker until blood is removed. Transfer embryos to 10ml syringe and push through into a 50ml Falcon tube. Weigh embryos and dilute in DMEM (1ml DMEM / 1g of minced embryos) and stir at +4°C over night. Add ice chilled hyaluronidase (4*10−5 g / 1g of minced embryos) and stir at +4°C for 1 hour. Ultracentrifuge lysates for 30 minutes at 46 000 g and filter sterilize (0.45 μm) the clear supernatant. Aliquot the chick embryo extract (~5–10 ml per tube) and store at −80°C (see Note 4).

  2. Due to the small starting volume, culture the dissociated tissue pieces in T25 flasks placed in upright position the first 2–3 days (see Note 10).

  3. After 2–3 days, add another 1–2ml of basic NC medium and add RA corresponding to new volume (V=3–4ml) (see Note 11). Always use a stripette to measure the current culture volume. In addition, upon adding fresh medium, slightly dissociate the spheres by pipetting them roughly ten times up and down towards the wall of the culture flask. Continue incubation with the T25 flasks in flat (‘lying’) position from hereafter (critically after a total volume of 3ml).

  4. Following another 2–3 days of incubation, add 3ml of basic NC medium and RA to new volume (V= 6–7ml) accordingly.

  5. Repeat step 4 following another 2–3 days of culture to reach total culture volume V=10ml. The user should strive to maintain this culture volume for continuous culturing (see Note 11).

  6. Add fresh retinoic acid (cranial and trunk) and BMP-4 (trunk) every 2 to 3 days (see Note 12).

  7. Crestospheres can be kept in culture for several weeks with maintained self-renewal capacity and expression of neural crest-associated genes (see Note 13).

3.3. Validation and Applications

  1. Before starting experiments with the newly established crestospheres, the end user should always validate the success of the procedure by assessing the expression of neural crest genes using in situ hybridization, Q-PCR, and/or immunostaining (see Note 14). Successful cultures are highly enriched in premigratory neural crest cells (expressing e.g. FoxD3), but markers for more differentiated neural crest progenitors (e.g. p75, HNK1), or low levels of central nervous system (CNS) neural genes (e.g. Sox2) may also be present (Fig. 2).

  2. Crestospheres provide a useful tool to study neural crest stemness and differentiation mechanisms and can be used for a myriad of in vitro applications, such as assays to validate cell proliferation, apoptosis, or migratory behavior (Figs. 1 and 2). Crestospheres are highly suitable for cell extraction for downstream gene expression studies, epigenetics, or protein-protein interactions. Gene expression in crestospheres can be modified by using retroviral transduction or chemical activators/inhibitors added to the culture medium. The developmental potential of crestospheres can also be tested in vivo by transplanting the spheres back into donor embryos to join the migratory neural crest stream (Fig. 2, and see Note 15). Single cell level analysis can be performed by either using confocal imaging or by imaging cryosections (see Note 16 for embedding and Note 17 for whole mount imaging).

Figure 2.

Figure 2.

Open in a new tab

Examples of techniques used for crestosphere validation and functional studies. A) Bright field microscopy. B) Crestospheres are highly enriched for neural crest cells as shown by in situ hybridization of FoxD3 and Sox10 expression in premigratory neural crest in the chicken embryonic dorsal neural tube and in crestosphres C) or by QPCR showing high FOXD3 and SOX10, whereas the expression levels of the neural stem cell gene SOX2 are low. D) Cryosections of crestospheres provide a tool for in depth examination of in situ hybridization, as exemplified by the expression of the EPAS1 gene, verifying that RNA expression is only localized at the outer edge of the sphere. E) Confocal high resolution imaging of a whole mount immunostained crestosphere by using an antibody to FoxD3. F) In vivo transplantation showing crestospheres after implantation into donor embryo at HH10. Scale bars represent 50μm. (A-C and E-F: Reprinted from “Crestospheres: Long-Term Maintenance of Multipotent, Premigratory Neural Crest Stem Cells”, Vol. 5/ Issue 4, Kerosuo L, Nie S, Bajpai R, Bronner ME, pp. 499–507, Copyright 2015, with permission from Elsevier.)

4. Notes

  1. Chick embryo development is monitored according to Hamburger Hamilton (HH) staging system, once determined by incubation at 37°C. This temperature might need to be adjusted by end user.

  2. Ringer’s solution: Dilute 40X Solution-1 and 40X Solution-2 to 1X in ddH2O. Add the solutions separately to water to avoid precipitation of calcium. Adjust pH to 7.4 (add ~1–2 drops of 1M NaOH) and filter sterilize using 0.22μm bottle top filters into autoclaved flasks.

  3. Prepare basic NC medium in advance, facilitating transfer of dissociated neural tubes to culture without delay in time. Basic NC media is stable for ~1 week at 4°C.

  4. Chick embryo extract (CEE) can be produced by end user or bought commercially. In both cases, run the extract through a 0.22μm bottle top filter, and store in 1–5ml aliquots at −20°C or −80°C.

  5. Retinoic acid (RA/ATRA) is resuspended in DMSO as a stock solution e.g. in 50μl aliquots in −20°C. Store and handle RA protected from light. Before use, thaw the aliquot in room temperature to prepare a ready to use dilution in PBS/culture medium, and importantly ensure at each occasion that the RA stock solution has retained its clear yellow/amber color and that ready-to-use RA also turns ‘yellowish’ in color. Ready-to-use RA should be vortexed immediately prior to addition to culture flasks. If the aliquot has precipitated, the color is more “neon-like” yellow, and the ready-to use suspension tube remains clear. Always prepare fresh diluted RA and discard left-overs after use.

  6. The use of low-adherent T25 culture flasks is crucial since attachment of cells may promote spontaneous differentiation.

  7. We recommend dissecting neural tubes from four (4) to six (6) embryos for successful establishment of cultures. Cranial-derived crestosphere cultures tend to grow and expand faster than trunk-derived cultures, thus aim for pools of at least five (5) embryos for trunk-derived cultures.

  8. The concentration of retinoic acid (RA) is of utmost importance and is known to play a role in neural crest specification in the neuroepithelium [11]. Crestospheres derived from different axial levels are dependent on different concentrations of RA (60nM and 180nM for cranial and trunk axial levels, respectively).

  9. Once neural tubes from all embryos to be pooled are dissected, proceed immediately to dissociation of sample as delays in the process may cause cell death and tissue degradation.

  10. For successful early expansion of crestosphere cultures, we recommend a small culture volume (V=2ml for cranial cultures with 6 embryos, and 1–2ml for trunk) and consequently incubating the T25 flask in upright position the first days following neural tube isolation. A small starting volume is critical for the success of the culture, too dilute of a cell concentration may kill off the culture.

  11. For gentle and step-wise expansion of crestospheres, small volumes of basic NC medium are added in the beginning of crestosphere culturing. This is continued until total culture volume reaches 10ml. A good rule of thumb is to add 25–50% more medium when you can clearly see an increase in crestosphere number by eye. A full flask with 10ml medium should contain hundreds of crestospheres. Monitor the cells daily, and add new medium more frequently, even daily, if the cultures grow fast, but as in the beginning, avoid too dilute conditions.

  12. Add new retinoic acid (RA) and BMP-4 every 2–3 days due to high turnover rates. Measure total culture volume by using a stripette and add factors accordingly.

  13. Crestospheres can be kept in culture for several weeks with maintained expression of neural crest-associated genes and self-renewal capacity. However, the longest time point examined is seven (7) weeks and at this time point the proliferation capacity is significantly decreased [7].

  14. Briefly, QPCR is performed by collecting 1–2 ml of sphere cultures in eppendorf tubes by centrifugation at maximum speed for 1 minute at +4°C. Remove supernatant and store pellets at −80°C until RNA extraction (method of choice by end user) and further QPCR analysis (method of choice by end user). Alternatively, allow crestospheres to sink down to the bottom of the tube (5–10 minutes), gently remove all culture medium and replace with respective RNA, DNA or protein extraction lysis buffer.

    Crestospheres can also be assessed for in situ hybridization according to protocols for whole embryos [12,13]. Unlike floating chicken embryos, crestospheres sink to the bottom of the tube during the hybridization step and washes, making the procedure straight forward. In order to not lose any crestospheres during washes, always let crestospheres sink to the bottom of the tube (5–10min), and leave ~100μl of the previous wash on the bottom of the tube before adding 1ml of the new wash.

  15. Transgenic GFP-positive chickens are available from Clemson Public Services Activities, Clemson University, SC, USA, and are useful for in vivo transplantation applications or for producing chimeras for clonal studies. They can also be transduced by using lentiviruses, with a typical transfection success rate of ~50%.

  16. Wild type, in situ hybridized or immunostained crestospheres can be embedded for cryosectioning. Briefly, crestosphere cultures are prepared by sucrose gradient (5% sucrose for 10 minutes followed by 15% sucrose for 2–4 hours at room temperature, allow spheres to collect at bottom of tube). Crestospheres are further primed in gelatin over night at 37°C. Embed crestospheres in gelatin using embedding molds. Make sure that the crestospheres do not sink to the bottom of the mold, but rather positions in the middle seen from all axes (position can be adjusted as gelatin is solidifying). Stiffened gelatin molds are snap-frozen by dipping the entire mold briefly but repeatedly in liquid nitrogen until complete samples are frozen. Store at −20°C for 5 minutes before transferring samples to Eppendorf tubes for long-term storage at −80°C before cryosectioning. Thickness for cryosections for immunostaining should range from 10–12μm, and 15–20μm for crestospheres after whole mount in situ hybridization (in order to get a strong signal).

  17. Crestospheres can be imaged whole mount or as cryosections by using regular microscopy (after in situ hybridization) or confocal microscopy (fluorescent signal). For high resolution imaging, use either of the following procedures: 1) prepare a “chamber” for the floating spheres in PBS by applying a vaseline wall (inject it from a 3ml syringe through a 25G5/8 needle) to make a small rectangle on a glass cover slip, pipette the spheres into the chamber, and cover it by using another glass cover slip; 2) transfer crestospheres to cavity slides and monitor for quicker but less detailed analysis. For lower magnification bright field imaging of pools of crestospheres after in situ hybridization, add a ~ 0.5 cm thick agarose (1% in H2O) layer on the bottom of the wells in a 24-well plate and let solidify. Place crestospheres in PBS-0.2% Tween on top of the agarose bed and image by using a dissection scope with a camera attached. Even the lightning by using a light source with adjustable, flexible arms.

5. Acknowledgement

This work was funded by the Swedish Childhood Cancer Foundation, Thelma Zoéga’s Foundation, Hans von Kantzow’s Foundation, the Mary Béve Foundation, The Royal Physiographic Society of Lund, Magnus Bergvalls stiftelse (to S.M.) and the Academy of Finland, Sigrid Juselius Foundation, and Ella and Georg Ehrnrooth’s Foundation (to L.K.).

6. References

  • 1.Simoes-Costa M, Bronner ME (2016) Reprogramming of avian neural crest axial identity and cell fate. Science 352 (6293):1570–1573. doi: 10.1126/science.aaf2729 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Dupin E, Calloni GW, Coelho-Aguiar JM, Le Douarin NM (2018) The issue of the multipotency of the neural crest cells. Dev Biol. doi: 10.1016/j.ydbio.2018.03.024 [DOI] [PubMed] [Google Scholar]
  • 3.Cohen AM, Konigsberg IR (1975) A clonal approach to the problem of neural crest determination. Dev Biol 46 (2):262–280 [DOI] [PubMed] [Google Scholar]
  • 4.Sieber-Blum M, Cohen AM (1980) Clonal analysis of quail neural crest cells: they are pluripotent and differentiate in vitro in the absence of noncrest cells. Dev Biol 80 (1):96–106 [DOI] [PubMed] [Google Scholar]
  • 5.Lahav R, Ziller C, Dupin E, Le Douarin NM (1996) Endothelin 3 promotes neural crest cell proliferation and mediates a vast increase in melanocyte number in culture. Proc Natl Acad Sci U S A 93 (9):3892–3897 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Trentin A, Glavieux Pardanaud C, Le Douarin N, Dupin E (2004) Self-renewal capacity is a widespread property of various types of neural crest precursor cells. Proc Natl Acad Sci U S A 101 (13):4495–4500 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kerosuo L, Nie S, Bajpai R, Bronner ME (2015) Crestospheres: Long-Term Maintenance of Multipotent, Premigratory Neural Crest Stem Cells. Stem Cell Reports 5 (4):499–507. doi: 10.1016/j.stemcr.2015.08.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kerosuo L, Bronner ME (2016) cMyc Regulates the Size of the Premigratory Neural Crest Stem Cell Pool. Cell Rep 17 (10):2648–2659. doi: 10.1016/j.celrep.2016.11.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Reynolds BA, Weiss S (1992) Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255 (5052):1707–1710 [DOI] [PubMed] [Google Scholar]
  • 10.Bajpai R, Chen D, Rada Iglesias A, Zhang J, Xiong Y, Helms J, Chang C-P, Zhao Y, Swigut T, Wysocka J (2010) CHD7 cooperates with PBAF to control multipotent neural crest formation. Nature 463 (7283):958–962 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Martinez-Morales P, Diez Del Corral R, Olivera-Martnez I, Quiroga A, Das R, Barbas J, Storey K, Morales A (2011) FGF and retinoic acid activity gradients control the timing of neural crest cell emigration in the trunk. The Journal of cell biology 194 (3):489–503 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kerosuo L, Bronner M (2014) Biphasic influence of Miz1 on neural crest development by regulating cell survival and apical adhesion complex formation in the developing neural tube. Mol Biol Cell 25 (3):347–355 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Acloque H, Wilkinson DG, Nieto MA (2008) Chapter 9 In Situ Hybridization Analysis of Chick Embryos in Whole-Mount and Tissue Sections. In: Marianne B-F (ed) Methods in Cell Biology, vol Volume 87. Academic Press, pp 169–185. doi: 10.1016/S0091-679X(08)00209-4 [DOI] [PubMed] [Google Scholar]
  • 14.Mohlin S, Kunttas E, Persson CU, Abdel-Haq R, Castillo A, Murko C, Bronner ME, Kerosuo L. Maintaining multipotent trunk neural crest stem cells as self-renewing crestospheres.Dev Biol. 2019. March 15;447(2):137–146. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES