Differentiation of Embryonic Stem Cells into Oligodendrocyte Precursors



We describe a small molecule-based protocol for differentiation of mouse embryonic stem cells into oligodendrocyte precursor cells (OPCs). This protocol generates Olig2+NG2+ OPCs with high efficiency by 30 days of differentiation. We also describe a method to generate "spiking" OPCs that can fire action potentials.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Jiang, P., Selvaraj, V., Deng, W. Differentiation of Embryonic Stem Cells into Oligodendrocyte Precursors. J. Vis. Exp. (39), e1960, doi:10.3791/1960 (2010).


Oligodendrocytes are the myelinating cells of the central nervous system. For regenerative cell therapy in demyelinating diseases, there is significant interest in deriving a pure population of lineage-committed oligodendrocyte precursor cells (OPCs) for transplantation. OPCs are characterized by the activity of the transcription factor Olig2 and surface expression of a proteoglycan NG2. Using the GFP-Olig2 (G-Olig2) mouse embryonic stem cell (mESC) reporter line, we optimized conditions for the differentiation of mESCs into GFP+Olig2+NG2+ OPCs. In our protocol, we first describe the generation of embryoid bodies (EBs) from mESCs. Second, we describe treatment of mESC-derived EBs with small molecules: (1) retinoic acid (RA) and (2) a sonic hedgehog (Shh) agonist purmorphamine (Pur) under defined culture conditions to direct EB differentiation into the oligodendroglial lineage. By this approach, OPCs can be obtained with high efficiency (>80%) in a time period of 30 days. Cells derived from mESCs in this protocol are phenotypically similar to OPCs derived from primary tissue culture. The mESC-derived OPCs do not show the spiking property described for a subpopulation of brain OPCs in situ. To study this electrophysiological property, we describe the generation of spiking mESC-derived OPCs by ectopically expressing NaV1.2 subunit. The spiking and nonspiking cells obtained from this protocol will help advance functional studies on the two subpopulations of OPCs.


Detailed procedures of generating oligodendrocyte precursor cells (OPCs) from GFP-Olig2 mouse embryonic stem cells (mESCs):

  1. Mouse ES cell line GFP-Olig2 (G-Olig2), is purchased from the American Type Culture Collection (ATCC). The mESCs are routinely passaged every 3 days onto an irradiated mouse embryonic fibroblast (MEF) feeder layer. The MEF cells are plated into 0.1% gelatin-coated six-well tissue culture plate at least one day before passaging the ESCs. The mESC culture medium is Dulbecco's modified Eagle's medium (DMEM) (GIBCO) supplemented with 20% fetal bovine serum (GIBCO), 2 mM L-glutamine, 1 mM sodium pyruvate, 0.1 mM β-mercaptoethanol/, 1% nonessential amino acids (NEAA), and 1,000 U/ml leukemia inhibitory factor.
  2. mES colonies are trypsinized (TrypLE, 5 min, 37°C) into single cells and suspended in knockout serum replacement (KSR) medium, then the cells are transferred to a Costar ultra-low attachment six-well plate (Corning) at a cell density of 50,000 cells/cm2. In these conditions, mESCs form embryoid bodies (EBs). KSR medium consists of -minimal essential medium (-MEM) supplemented with 20% KSR, 1 mM sodium pyruvate, 1% NEAA, and 0.1 mM β-mercaptoethanol/.
  3. From day 4 to day 7, retinoic acid (RA, 0.2 M) and purmorphamine (Pur, 1 M) are included as shown in Fig. 1A in KSR or N2 medium. The N2 medium is -MEM containing 1X N2 supplement, 1 mM sodium pyruvate, 1% NEAA, and 0.1 mM β-mercaptoethanol/. The medium is changed everyday.
  4. At day 8, EBs are disaggregated using TrypLE (5min, 37°C) and plated on 0.01% polyornithine-coated dishes in OPC medium that consists of N2 medium and fibroblast growth factor-2 (FGF-2, 20 ng/ml). The medium is changed every two days.
  5. The cells are trypsinized (TrypLE, 3 min, 37°C) and replated approximately every week when they become confluent. At day 30, over 80% of the cells show GFP/Olig2 expression and are NG2-positive, characteristic of OPCs.
  6. To generate spiking OPCs, the BacMam Na+ channel kit is used to introduce Nav1.2 subunit into these mESC-derived OPCs. The BacMam reagent (1 ml, component A) is mixed with PBS to a final volume of 5 ml. Then, the mixed BacMam reagent is added into mESC-derived OPCs culture in a 60 mm culture dish (at 50-70% confluence) prior to rinsing with PBS. After 2 hr incubation, the mixed BacMam reagent is removed and replaced with OPC medium supplemented with 1X enhancer. The enhancer is the component B reconstituted in DMSO (component C). Next, after additional 2-hr incubation the medium with enhancer is removed and replaced with complete OPC medium. The cells are assayed 24 hrs later.

Representative Results:

Following the differentiation protocol shown in Fig. 1A, G-Olig2 mESCs were initially suspended in KSR medium and for 4 days to form embryoid bodies (EBs) (Fig. 1B). Then, we sequentially treated the EBs with RA and Pur. The EBs did not show any GFP fluorescence at Day 4 and expressed strong GFP fluorescence at D8 (Fig. 1B). At this time point, the EBs were trypsinized and plated in N2 medium supplemented with growth factors FGF-2. After further culture for 22 days (D30), the percentage of GFP+ cell reached 80.3 ± 0.6 % according to our flow cytometry results. As shown in Fig. 1C, GFP+ cells overlapped with NG2 staining consistently (96.4 ± 1.3% of GFP positive cells were NG2 positive, and 94.8 ± 1.5% of NG2 positive cells were GFP positive). Thus, these cells express GFP/Olig2 and NG2, defining them as OPCs.

The mESC-derived OPCs (76 cells) failed to fire action potentials upon depolarization (Fig. 2A). After transducing with a baculovirus carrying the Nav1.2 subunit, these mESC-derived OPCs (94.1%, 16 of 17 cells) could be classified as spiking (Fig. 2B).

Figure 1
Figure 1. Differentiation of GOlig-mESC into OPCs. (A) Scheme showing the protocol of the embryoid body (EB)-based and small molecule-driven differentiation. At D8, the EBs were disaggregated and plated. The cells were passaged once per week when they became confluent. (B) D8, but not D4 EBs, showed GFP expression. (C) Immunostaining of NG2 (red) was consistent with GFP (green) expression at D30. DAPI (blue) was used to identify the nuclei. Scale bar: 20 mm.

Figure 2
Figure 2. Generation of spiking OPCs from nonspiking mESC-derived OPCs by virus-mediated Nav channel expression.  (A) The membrane potential was held at -60 mV and mESC-derived OPCs failed to fire action potentials, even when the cell was depolarized to 0 mV. (B) An example of mESC-derived OPC firing action potential (highlighted in red) after the viral transduction.


Embryonic stem cells (ESCs), isolated from a blastocyst embryo, can differentiate into all cell lineages of the organism 3, 4, providing an in vitro model system for studying early mammalian development, including oligodendrocyte specification. mESCs have been shown to differentiate into oligodendrocyte precursor cells (OPCs) with the treatment of sonic hedgehog (Shh) 5. Moreover, the Shh-induced OPC differentiation from mESCs retains the correct timing observed in embryonic development 6. Hence, the nature of in vitro OPC differentiation from mESCs is considered to be consistent with what has been learned from in vivo development. Here, using the small molecules RA and the Shh agonist Pur, we successfully differentiated the G-Olig2 mESCs into GFP+Olig2+NG2+ OPCs with high efficiency.

OPCs, characterized by the expression of the proteoglycan NG2 7 and the helix-loop-helix transcription factor Olig2 8, generate oligodendrocytes in the developing and mature CNS, where they comprise a significant percentage (~ 5%) of the total cells and are the main proliferating cell type 9. For nearly two decades researchers have demonstrated NaV channels are expressed in a subpopulation of OPCs and can be activated upon depolarization 10, 11. Moreover, a recent striking observation9 was that in in situ rat CNS white matter, OPCs (~ 50%) generated action potentials when depolarized depending upon the expression of voltage-gated sodium (NaV) channels, and thus could be subdivided into spiking and nonspiking subpopulations. However, the functions of these spiking properties are still largely unknown. We found that, electrophysiologically, the mESC-derived OPCs differentiated with the Shh-dependent protocol, were not the same as the in situ brain OPCs. After introducing subunit NaV1.2, these silent mESC-derived OPCs were capable of spiking.

Thus, the spiking/nonspiking mESC-derived OPCs and differentiation protocol described here may facilitate (1) studying the functional differences between spiking and nonspiking OPCs, (2) screening new factors that could promote sodium channel expression in the mESC-derived OPCs, (3) developing and optimizing the differentiation protocol of OPCs from human ESC or induced pluripotent stem cells (iPSCs).


No conflicts of interest declared.


This work was in part supported by grants to W.D. from National Institutes of Health (RO1 NS059043 and RO1 ES015988), National Multiple Sclerosis Society, Roche Foundation for Anemia Research, Feldstein Medical Foundation, and the Shriners Hospitals for Children.

We would like to thank Dr. David Pleasure and Jennifer Plane for suggestions. We declare no competing interests related to this article.


Name Company Catalog Number Comments
MEFs GlobalStem GSC-6001G
TrypLE Express Invitrogen 12604
Retinoic acid Sigma-Aldrich R-2625
Purmorphamine Cayman Chemical 10009634
FGF-2 EMD Millipore GF003
BacMam NaV 1.2 Invitrogen B10341



  1. Xian, H., Gottlieb, D. I. Dividing Olig2-expressing progenitor cells derived from ES cells. Glia. 47, (1), 88-101 (2004).
  2. Clair, A., Gottlieb, D. I. A subset of ES-cell-derived neural cells marked by gene targeting. Stem Cells. 21, (1), 41-49 (2003).
  3. Evans, M. J., Kaufman, M. H. Establishment in culture of pluripotential cells from mouse embryos. Nature. 292, 154-156 (1981).
  4. Thomson, J. A. Embryonic stem cell lines derived from human blastocysts. Science. 282, 1145-1147 (1998).
  5. Shin, S., Xue, H., Mattson, M. P., Rao, M. S. Stage-dependent Olig2 expression in motor neurons and oligodendrocytes differentiated from embryonic stem cells. Stem Cells Dev. 16, 131-141 (2007).
  6. Billon, N., Jolicoeur, C., Ying, Q. L., Smith, A., Raff, M. Normal timing of oligodendrocyte development from genetically engineered, lineage-selectable mouse ES cells. J Cell Sci. 115, 3657-3665 (2002).
  7. Nishiyama, A., Lin, X. H., Giese, N., Heldin, C. H., Stallcup, W. B. Co-localization of NG2 proteoglycan and PDGF alpha-receptor on O2A progenitor cells in the developing rat brain. J Neurosci Res. 43, 299-314 (1996).
  8. Ligon, K. L. Development of NG2 neural progenitor cells requires Olig gene function. Proc Natl Acad Sci U S A. 103, 7853-7858 (2006).
  9. Dawson, M. R., Polito, A., Levine, J. M., Reynolds, R. NG2-expressing glial progenitor cells: an abundant and widespread population of cycling cells in the adult rat CNS. Mol Cell Neurosci. 24, 476-488 (2003).
  10. Chittajallu, R., Aguirre, A., Gallo, V. NG2-positive cells in the mouse white and grey matter display distinct physiological properties. J Physiol. 561, 109-122 (2004).
  11. Karadottir, R., Hamilton, N. B., Bakiri, Y., Attwell, D. Spiking and nonspiking classes of oligodendrocyte precursor glia in CNS white matter. Nat Neurosci. 11, 450-456 (2008).



  1. Your protocol says RA was used at 0.² M and purmorphamine was used at 1 M. Those concentrations are surprisingly high. Can you verify the correct final concentrations of each in the differentiation medium?

    Posted by: patrick m.
    May 16, 2012 - 9:58 PM
  2. The concentration unit should be uM.

    Posted by: Wenbin D.
    May 16, 2012 - 10:16 PM
  3. Yes, RA 0.² microM and Pur 1 microM.

    Posted by: Peng J.
    May 17, 2012 - 12:47 AM
  4. you applied purmorphamine in your protocol. In your opinion whether we could use SHH instead of it or not? If it could be possible what concentration we should use?

    Posted by: maryam p.
    June 7, 2012 - 8:52 AM
  5. yes, you can use SHH. Shh at 30 nM should work.

    Posted by: Peng J.
    June 7, 2012 - 1:29 PM
  6. Thanks for your attention,
    could we passage this precursor cells after 30 days?how much?

    Posted by: maryam p.
    June 7, 2012 - 2:10 PM
  7. Thanks for your attention,
    could we passage this precursor cells after 30 days?how much?

    Posted by: maryam p.
    June 8, 2012 - 3:45 PM
  8. Yes you can passage when cells become confluent.

    Posted by: Peng J.
    June 8, 2012 - 5:05 PM
  9. you use KSR medium consists of -minimal essential medium (-MEM) for EBs formation,what's the meaning of -minimal essential medium? is DMEM?

    Posted by: maryam p.
    June 9, 2012 - 3:08 PM
  10. It is alpha-MEM from Invitrogen.
    If you have further questions, you can send email to me. My email address is pjiang@ucdavis.edu

    Posted by: Peng J.
    June 9, 2012 - 9:37 PM
  11. would hES cells have the same results? would there be technical/timing issues involved in running an experiment with hES cells rather than mES cells?

    Posted by: Ryan M.
    July 19, 2012 - 7:16 AM
  12. There are many differences between differentiating hESCs and mESCs. Please see our publication in Nature Protocols. http://www.ncbi.nlm.nih.gov/pubmed/²15²79²1

    Posted by: Peng J.
    July 19, 2012 - 12:46 PM
  13. Thank you. I'm a high school researcher, so I don't have quite the knowledge base. Your video is helping me along greatly

    Posted by: Ryan M.
    July 19, 2012 - 7:15 PM
  14. Can the OPCs generated by this method be differentiated into type-² astrocytes? Are these OPCs phenotypically similar to O²A cells?

    Posted by: Anonymous
    October 5, 2012 - 5:43 PM
  15. Yes, the mESC-OPCs can generate type-1 and type-² astrocytes in vitro. For details, you can check this link. http://www.sciencedirect.com/science/article/pii/S030439401²0089²0

    Posted by: Peng J.
    October 5, 2012 - 8:35 PM

Post a Question / Comment / Request

You must be signed in to post a comment. Please or create an account.

Usage Statistics