Transgenic Enrichment of Mouse Embryonic Stem Cell-derived Progenitor Motor Neurons

Abstract

Embryonic stem cells (ESCs) hold great potential for replacing neurons following injury or disease. The therapeutic and diagnostic potential of ESCs may be hindered by heterogeneity in ESC-derived populations. Drug selection has been used to purify ESC-derived cardiomyocytes and endothelial cells but has not been applied to specific neural lineages. In this study we investigated positive selection of progenitor motor neurons (pMNs) through transgenic expression of the puromycin resistance enzyme, puromycin N-acetyl-transferase (PAC), under the Olig2 promoter. The protein-coding region in one allele of Olig2 was replaced with PAC to generate the P-Olig2 cell line. This cell line provided specific puromycin resistance in cells that express Olig2, while Olig2⁻ cells were killed by puromycin. Positive selection significantly enriched populations of Olig2⁺ pMNs. Committed motoneurons (MNs) expressing Hb9, a common progeny of pMNs, were also enriched by the end of the selection period. Selected cells remained viable and differentiated into mature cholinergic MNs and oligodendrocyte precursor cells. Drug resistance may provide a scalable and inexpensive method for enriching desired neural cell types for use in research applications.

1. Introduction

In most neurological disorders, neurogenesis is insufficient to replenish lost neuronal populations. Endogenous stem cell populations are hindered by limited numbers, variable proliferation in response to disease, and in some cases, differentiation into glia rather than neurons [1-4]. Embryonic stems cells (ESCs) can be differentiated into specific neuronal subtypes and may be useful for cell replacement strategies in the central nervous system [5]. Transplantation of ESC-derived dopaminergic neurons and cholinergic motoneurons (MNs) has been shown to promote partial recovery from Parkinson’s-like symptoms and spinal cord injury, in rodent models [6, 7]. Heterogeneous populations arising from differentiation of ESCs, however, currently limit the efficacy of such treatments [8]. Strategies for controlled differentiation of ESCs and the subsequent enrichment ESC-derived cells types are therefore critical to the development of ESC-based therapies and diagnostic screening tools.

Directed differentiation of ESCs into spinal MNs can be achieved following exposure to retinoic acid (RA) and sonic hedgehog (Shh) [9, 10]. During this process, ESCs first differentiate into progenitor motor neurons (pMNs) expressing the basic helix-loop-helix transcription factor Olig2 [11, 12]. These cells can commit to the MN fate by downregulating Olig2 and expressing the homeodomain (HD) transcription factors Islet 1 (Isl1) and Hb9, also known as Mnx1 [11-14]. Despite optimization, differentiation protocols for pMNs result in a heterogeneous population of cells including other ventral spinal progenitor cells [10]. Hb9⁺-committed MNs compose only 15-50% of the total culture after differentiation of ESCs [9, 15]. Low-purity cultures give rise to multiple types of spinal interneurons, therefore subsequent enrichment may be necessary [15].

Greater pMN purity can be obtained by fluorescence-activated cell sorting (FACS) of a transgenic ESC line that expresses GFP under the Olig2 gene regulatory elements (GRE) [16-18]. This method, however, requires expensive equipment and must be performed at a centralized facility, risking contamination. Gradient centrifugation can enrich spinal MNs from the mouse embryonic lumbar spinal cord and human ESCs, but has not been optimized for mouse ESC-derived MNs [19, 20]. Transgenic selection may provide a low-cost alternative and can be performed directly in the culture dish.

Puromycin resistance through expression of the enzyme puromycin N-acetyl-transferase (PAC) has been shown to allow enrichment of ESC-derived cardiomyocytes and endothelial cells in transgenic lines [21-24], but has not been used to enrich specific neural populations.

In this study, we investigated whether transgenic selection could help to enrich low-purity populations that commonly result from pMN differentiation protocols. We generated a new heterozygous “knock in” mouse ESC line (P-Olig2) where the protein-coding region in one allele of Olig2 was replaced with PAC, allowing for positive selection of Olig2⁺ pMNs during the differentiation. Olig2 expression was analyzed during directed differentiation of ESCs into pMNs using the Shh signaling agonist, purmorphamine [25, 26]. Puromycin-treated cells were assessed for expression of pMN-specific markers and differentiation into pMN progeny, including MNs and oligodendrocytes. This study demonstrates the first use of puromycin resistance for positive selection of a specific population of neural progenitor cells.

2. Results

2.1 Olig2 expression during differentiation of ESCs

To determine the effect of Shh signaling levels on directed differentiation of ESCs into pMNs, we analyzed mRNA levels in response to increasing concentrations of purmorphamine, a Shh agonist, using quantitative real time (RT)-PCR. ESCs were exposed to 2 μM retinoic acid (RA) and 250 nM, 500 nM, or 1 μM purmorphamine. Relative mRNA levels were analyzed at the end of the 2⁻/4⁺ differentiation protocol and were compared to control cells that did not receive RA or purmorphamine (n=3 for all conditions). Increasing the purmorphamine concentration from 250 nM to 1 μM led to downregulation of Dbx2 and Irx3, two transcription factors found in p1 and p2 progenitor (more dorsal) domains, respectively (Figure 1A-B). The mRNA levels for Pax6, which is expressed in the p1, p2, and pMN domains, did not change with concentration. Nkx2.2 mRNA levels were too low for detection even at the highest concentration of purmorphamine (data not shown). Olig2 expression significantly increased with exposure to 1 μM purmorphamine compared to 250 nM and 500 nM purmorphamine (Figure 1C). HD transcription factors Isl1 and Hb9 expressed during commitment of pMNs to the MN fate were upregulated with 1 μM purmorphamine, similar to Olig2. Finally, 1 μM purmorphamine led to an increase in mRNA for choline acetyltransferase (ChAT), an enzyme found specifically in mature MNs.

Figure 1.

Transcription factor expression during directed differentiation of ESCs. (A): Schematic of transcription factors expressed in spinal progenitor domains. Ventral-to-dorsal gradient of sonic hedgehog (Shh) and relative position of spinal progenitor domains in the ventral neural tube are shown on the left. The pattern of transcription factor expression in each domain is shown on the right. (B): Expression of p1 and p2 progenitor genes in ESCs differentiated with the Shh signaling agonist, purmorphamine, and retinoic acid (RA) using a 2⁻/4⁺ differentiation protocol. Data are expressed as the fold difference in mRNA levels compared to ESCs differentiated without purmorphamine and RA. (C): Expression of pMN and MN specific genes in differentiated ESCs. * indicates p < 0.05 for that gene compared to ESCs exposed to 250 nM purmorphamine. # indicates p < 0.05 for that gene compared to ESCs exposed to 500 nM purmorphamine. Abbreviations: FP, floor plate; ChAT, choline acetyl transferase; Isl1, Islet 1.

2.2 Heterogeneity in differentiated cultures

To characterize the heterogeneity of the cell population resulting from differentiation of ESCs, we utilized a transgenic ESC line expressing GFP under the Olig2 GRE, G-Olig2 [17]. GFP fluorescence can persist for several days after transcription, allowing for identification of pMNs and their recent progeny during differentiation. By visual inspection of GFP fluorescence, pMNs could be easily separated from other spinal progenitor cells that differentiate into spinal interneurons rather than MNs.

G-Olig2 ESCs were differentiated using 1 μM purmorphamine and 2 μM RA and analyzed by flow cytometry. At the end of the 2⁻/4⁺ differentiation protocol, 61.6 ± 4.5% of cells expressed GFP (n = 3). This percentage was approximately 3-fold higher than expression of Olig2 found with traditional antibody staining, suggesting that GFP fluorescence persists in recent pMN progeny as expected. GFP⁻ cells were consistently found migrating away from EBs and displayed a broad flat morphology typical of astrocytes or astrocyte precursor cells (ASPs) (Figure 2 D-F; white arrows). These cells may originate from the p2 progenitor domain that gives rise to ASPs in vivo [31]. Additional GFP⁻ cells were present within EBs with a similar morphology to Olig2⁺ pMNs and may represent progenitor cells from adjacent progenitor domains (white asterisk).

Figure 2.

Heterogeneity following directed differentiated of G-Olig2 ESCs. (A): Phase contrast image of differentiated G-Olig2 ESCs following exposure to purmorphamine and RA. (B): Expression of green fluorescent protein (GFP) in (A) showing cells that have expressed Olig2 during differentiation. (C): Corresponding Hoechst staining showing cell nuclei. Scale bars = 100 μM. (D-F) Close inspection of GFP⁻ cells (white arrows) with astrocyte-like morphology that did not differentiate into pMNs expressing Olig2. Additional GFP⁻ cells with progenitor-like morphology were found within EBs (white asterisk). Scale bars = 50 μM.

2.3 Generating the P-Olig2 Cell Line

The P-Olig2 cell line was generated using a targeting vector with a resistance cassette in the open reading frame of the Olig2 gene surrounded by two regions homologous to the Olig2 locus. RW4 ESCs were electroporated with the P-Olig2 targeting vector and homologous recombination occurred as illustrated in (Figure 3A). To confirm targeted insertion, novel junctions were detected using short arm junction PCR. Successful integration resulted in a 2.1 kb fragment spanning from inside the targeting construct into neighboring genomic DNA (Figure 3B). The addition of the resistance cassette increased the distance between Hind III sites within the Olig2 locus from 4.8 kb to 6.7 kb as observed by Southern analysis (Figure 3C). Only the 4.8kb band was observed in control RW4 cells while both bands were observed in the P-Olig2 cell line. The appearance of both bands confirms that only one of the two Olig2 loci was targeted. Following insertion, the floxed PGK-neo cassette was excised with Cre recombinase as confirmed using PCR by a new 350 bp band (Figure 3D). The final cell line contained a promoter-less PAC cassette driven by the Olig2 GRE (Figure 4A).

Figure 3.

Knock-in to replace Olig2 ORF with PAC gene. (A): Olig2 gene is a schematic of the Olig2 gene. The smaller black boxes represent the two exons of the Olig2 gene with the ORF (large black box) located in the second exon. The 5′ probe and H (HindIII) sites used for Southern blots are indicated. P-Olig2 vector shows the targeting vector with the PAC and floxed neo cassette in place of the ORF. P-Olig2-neo is the predicted knock-in product with the predicted 6.7Kb HindIII fragment and the 2.1 Kb junction PCR product shown. P-Olig2 shows the engineered gene after Cre-excision of the floxed neo cassette. (B): Junction PCR of control RW4 cells and targeted cells. Amplified DNA is absent in the RW4 cells but present in the targeted (P-Olig2) cells. (C): Southern blots-RW4 parental cells have native 4.8 Kb HindIII band; targeted (POlig2) cells have additional 6.7 Kb band. (D) Cre-excision of neo cassette-PCR reactions across neo cassette show amplified DNA in Cre⁺ cells but not in untreated (Cre⁻) cells.

Figure 4.

PAC and Olig2 expression in the P-Olig2 cell line (A): Schematic of the PAC cassette driven by the native Olig2 GRE. (B): Olig2 and PAC mRNA expression in P-Olig2 and RW4 cell lines. Expression was analyzed in ESCs, ESCs aggregated into embryoid bodies (EBs), and ESCs differentiated into pMNs using a 2⁻/4⁺ treatment protocol. (C): Live/dead assay showing live cells (green) and dead cells (red) following exposure of undifferentiated P-Olig2 and RW4 ESCs to puromycin for 48 hours. (D): Live/dead assay for differentiated P-Olig2 and RW4 EBs selected with puromycin for 48 hours. Fluorescent cell debris was observed in the RW4 group whereas the P-Olig2 group contained whole viable cells.

To determine the specificity of PAC expression, P-Olig2 and RW4 ESCs were differentiated into pMNs, and the expression of Olig2 and PAC mRNA was assessed using PCR. Olig2 and PAC mRNA was not observed or present at very low levels in ESCs and ESCs aggregated into EBs (Figure 4B). Following the 2⁻/4⁺ differentiation protocol, Olig2 mRNA levels were elevated in both P-Olig2 and RW4 cell lines. PAC mRNA, however, was specific to the P-Olig2 cell line. Specificity of PAC expression in the P-Olig2 cell line was confirmed by sensitivity to puromycin. Few P-Olig2 ESCs survived following exposure to puromycin for 48 hrs (Figure 4C) since the Olig2 gene is off or expressed at very low levels. In contrast, many viable cells were observed in cultures of P-Olig2 ESCs differentiated into pMNs, which express Olig2 (Figure 4D). Widespread cell death was still observed in the latter group, suggesting the presence of Olig2⁻ cells that do not express PAC and are sensitive to puromycin.

2.4 Enrichment of pMNs using the P-Olig2 Cell Line

To determine whether transgenic selection enriched the cell population for pMNs and committed MNs, P-Olig2 cells were differentiated using 2 μM RA and 1.5 μM purmorphamine with a 2⁻/4⁺ differentiation protocol. Puromycin was added during the last two days to select for Olig2⁺ pMNs (Figure 5A). Control groups consisted of RW4 and P-Olig2 cells not receiving puromycin. Similar to G-Olig2 cell cultures, broad Olig2⁻ cells were observed migrating out of the EB in the control P-Olig2 group (Figure 5B). In cultures with puromycin, these cells were not present. Cellular debris localizing to the same region suggests that puromycin-induced cell death occurred in Olig2⁻ cells. The majority of surviving cells had a progenitor-like morphology and expressed Olig2 or Hb9.

Figure 5.

Positive selection of P-Olig2 ESCs differentiated using a 2⁻/4⁺ differentiation protocol. (A): Schematic showing 2⁻/4⁺ differentiation protocol of ESCs. Puromycin was added from day 4 and remained till day 6. (B): Immunocytochemistry analysis of P-Olig2 ESCs following differentiation and puromycin selection. Olig2⁺ and Hb9⁺ cells compose a small fraction of the entire population in the absence of puromycin. Following puromycin treatment, the majority of cells expressed either Olig2 or Hb9. (C): Nestin expression in control and selected P-Olig2 cultures. The majority of cells express nestin. (D): Flow cytometry histograms and gating. Solid black histograms represent secondary antibody controls. (E): Flow cytometry analysis of RW4, non-selected P-Olig2, and puromycin selected P-Olig2 ESCs following differentiation. * indicates p < 0.05 for that marker compared to unselected P-Olig2 group. Scale bars = 100 μM.

To assess cell differentiation in control and selected cultures, cells were stained at the end of the differentiation protocol with phenotype-specific markers and analyzed using flow cytometry. Puromycin selection did not affect the distribution of Nestin⁺ cells, suggesting that the majority of cells were neural cells. The percentage of Olig2⁺ cells increased significantly from 21.9 ± 5.9% to 48.7 ± 7.3% when selected with puromycin (Figure 5C). Hb9⁺ populations were enriched from 20.6 ± 7.9% to 58.5 ± 1.5%. Similar enrichment was observed for Isl1. No significant differences were found for any marker between the RW4 and control P-Olig2 (non-selected) groups.

2.5 Undifferentiated Stem Cells

To determine the effect of puromycin exposure on Oct4⁺ undifferentiated stem cells, cultures were fixed and analyzed following immunocytochemistry (ICC). The overall percentage of cells expressing the pluripotent stem cell marker Oct4 following the 2⁻/4⁺ differentiation was low (<1%) in control and selected groups (data not shown). The fraction of 2⁻/4⁺ EBs containing at least one Oct4⁺ nuclei was compared between unselected and selected P-Olig2 pMNs on day 6. Oct4⁺ cells were commonly found in small groups and were limited to unselected P-Olig2 cultures (Figure 6A). Approximately 10% of EBs contained undifferentiated stem cells in the absence of puromycin (Figure 6B). Oct4⁺ nuclei were not observed in any of the cultures selected with puromycin.

Figure 6.

Oct4 expression in P-Olig2 cells following the 2⁻/4⁺ differentiation protocol. (A): A small group of Oct4⁺ nuclei in an unselected, control P-Olig2 EB. (B): Fraction of EBs containing Oct4⁺ nuclei in control and selected P-Olig2 cultures. * indicates p < 0.05 compared to unselected P-Olig2 group

2.6 Extended Differentiation of pMNs

To determine whether targeted replacement of the Olig2 gene impacted the ability of ESC-derived pMNs to terminally differentiate, we cultured puromycin selected P-Olig2 cells for two weeks on laminin-coated wells. Following differentiation, cells were assessed for expression of ChAT and neurofilament (NF) using ICC to identify mature MNs. Cultures were also stained for oligodendrocyte marker 4 (O4) and RIP to identify oligodendrocytes, another common progeny of pMNs. Groups of ChAT⁺/NF⁺ neurons were abundant throughout the culture (Figure 7). Oligodendrocytes expressing O4 and RIP were also present (Figure 7). These results demonstrate that cells surviving the puromycin selection differentiate into neurons and oligodendrocytes, the expected progeny of Olig2⁺ pMNs.

Figure 7.

Immunocytochemistry performed on selected pMNs differentiated for two weeks on laminin. (A): ChAT and respective (B): NF expression in differentiated cultures showing mature cholinergic motoneurons. Differentiation also led to expression of (C): O4 and (D): RIP showing immature and mature oligodendrocytes respectively. Scale bars = 100 μM. Abbreviations: ChAT, choline acetyl transferase; NF, neurofilament; O4, oligodendrocyte marker 4.

3. Discussion

Current protocols for differentiation of mouse ESCs are often hindered by low efficiencies. Directed differentiation may be lead to heterogeneous ESC-derived populations that must be further purified to obtain the desired lineages prior to cell culture studies or transplantation. In this study, we demonstrate that positive selection of Olig2⁺ pMNs through transgenic expression of the puromycin resistance enzyme PAC can provide a simple method for enrichment of pMNs.

Purmorphamine exhibited a dose-dependent effect on pMN gene expression during directed differentiation of G-Olig2 ESCs. Greater concentrations of purmorphamine led to a significant increase in Olig2 expression. Conversely, expression of Dbx2 and Irx3 (more dorsal transcription factors) were reduced with increasing concentrations of purmorphamine. Nkx2.2, which is expressed in the more ventral p3 progenitor domain, was not detected in any of the conditions tested. Based on the gene expression data, directed differentiation appears to favor pMNs at the highest concentration of purmorphamine tested. However, this condition still resulted in a mixed cell population with nearly 40% of cells not expressing Olig2. This undesired population may include cells from neighboring progenitor domains that have been previously observed following directed differentiation of ESCs into pMNs [10]. Differentiation of ESCs into cells from multiple spinal progenitor domains is potentially due to overlapping dependency on Shh signaling in the ventral neural tube or our inability to precisely control localized concentration of purmorphamine over the duration of the experiment. Furthermore, variations in the responsiveness of each cell to Shh signaling may attribute to heterogeneity.

Positive selection of cardiomyocytes and endothelial cells through puromycin resistance has been previously shown using randomly inserted resistance cassettes containing a cell-type specific promoter [21-23]. By knocking in PAC expression, we preserve regulatory mechanisms for the native Olig2 gene. Expression of PAC in the final P-Olig2 cell line recapitulated expression of Olig2 in the native allele. Specificity of the PAC cassette driven by the native Olig2 GRE was shown by puromycin sensitivity. Cell death was induced in puromycin treated P-Olig2 ESCs within 48 hours which is similar to the time-course previously described for puromycin-induced cell death [32]. Only when P-Olig2 ESCs were differentiated into pMNs using the 2⁻/4⁺ differentiation protocol did cells remain viable following puromycin treatment. Cell death was still observed in differentiated ESC cultures suggesting that only cells having expressed Olig2 show resistance. Consistent with this hypothesis, puromycin treatment removed all broad flat cells previously identified as Olig2⁻. In addition, puromycin killed all Olig2⁻/Oct4⁺ undifferentiated stem cells. The majority of remaining viable cells were positive for the transcription factors Olig2 or Hb9. These results demonstrate selective resistance in pMNs and their progeny.

Differentiation of non-selected P-Olig2 ESCs was similar to RW4 ESCs for all markers tested in this study. Substituting PAC for one copy of the Olig2 gene did not appear to alter differentiation of P-Olig2 ESCs into pMNs or commitment of pMNs to the MN fate. Following puromycin treatment, Olig2⁺ pMNs were significantly enriched. High purity pMNs cultures were not obtained in this study; perhaps this is due to the persistence of the PAC enzyme in committed Hb9⁺ MNs. Committed MNs were enriched nearly 3-fold and constituted the majority of cells at the end of the selection. This population may become further enriched as additional Olig2⁺ pMNs commit to the MN fate. The distribution of Nestin⁺ cells was not affected by puromycin treatment, suggesting that the majority Olig2⁻ cells killed by puromycin were neural lineage cells. These cells could be other spinal progenitor cells that express nestin.

The long-term effects of PAC knock-in and puromycin exposure were assessed through differentiation of puromycin-treated pMNs on laminin. Following a two week differentiation period, ChAT⁺/NF⁺ MNs and O4⁺/RIP⁺ immature oligodendrocytes were observed. These results demonstrate long-term viability of enriched pMNs and committed MNs, suggesting no persistent effects of puromycin exposure. Oligodendrocyte development from pMNs is dependent on continuous Olig2 expression [17], and does not appear to be disturbed by the PAC knock-in. Both MNs and oligodendrocytes have been transplanted for treatment of SCI [33, 34] and are potential target populations for cell replacement strategies in other neurological disorders including amyotrophic lateral sclerosis and multiple sclerosis. Additional enrichment strategies can be employed to purify individual populations of MNs or oligodendrocytes.

4. Conclusions

Many studies in developmental biology, drug screening, and regenerative medicine can benefit from high purity ESC-derived cell populations. This study demonstrates the use of puromycin resistance to select for a well-defined set of neural progenitor cells following directed differentiation of ESCs. The methods utilized in this study can be applied to other neural cell types to generate high-purity populations that can improve the use of ESC-derived populations in research.

5. Materials and Methods

5.1 Embryonic Stem Cell Culture

The RW4 ESC line was used to generate and characterize the P-Olig2 cell line. Olig2 expression was characterized using the G-Olig2 ESC line that expresses GFP under the control of the Olig2 GRE [16-18]. ESCs were grown in complete media consisting of Dulbecco’s modified Eagle’s Medium (DMEM; Invitrogen, Carlsbad, CA, http://www.invitrogen.com) supplemented with 10% newborn calf serum (Invitrogen), 10% fetal bovine serum (Invitrogen), 10 μM thymidine (Sigma, St. Louis, MO, http://www.sigmaaldrich.com), and 30 μM of each of the following nucleosides: adenosine, cytosine, guanosine, and uridine (Sigma). Cells were passaged at a 1:5 ratio every 2 days and seeded on a new T25 flask coated with a 0.1% gelatin solution (Sigma). After seeding, 1000 U/ml leukemia inhibitory factor (LIF; Millipore, Billerica, MA, http://www.millipore.com) and 100 μM β-mercaptoethanol (BME; Invitrogen) were added to the media to maintain the undifferentiated state of the ESCs without the need for a feeder cell layer.

5.2 Generation of P-Olig2 ESCs

The P-Olig2 cell line was generated from the RW4 line. Approximately 1×10⁷ RW4 ESCs were resuspended in electroporation buffer with 10-15 μg of ScaI-linearized P-Olig2 targeting vector. The targeting vector was constructed from a Gateway-compatible plasmid (pStartK) incorporating the Olig2 locus with the Olig2 open reading frame replaced by a dual resistance cassette consisting of from 5′ to 3′: Asc1 site, Kozak sequence, puromycin cassette with bgh polyA signal (PKO-Select Puro, Agilent Genomics, Santa Clara, CA, http://www.genomic.agilent.com), floxed phosphoglycerate kinase I promoter driving the neomycin phosphotransferase gene (PGK-neo) with bgh polyA signal, and AscI site [27-29]. Cells were electroporated using an Amaxa nucleofector II (Lonza, Basel, Switzerland, http://www.lonzabio.com) at 0.23kV and 960μF in a 0.4 cm cuvette (Bio-Rad, Hercules, CA, http://www.bio-rad.com). Following electroporation, cells were seeded on gelatin coated 10cm dishes for 24 hours then dosed with G418 (200 μg/ml, Invitrogen) and 1-(2-Deoxy-2-fluoro-β-D-arabinofuranosyl)-5-iodouracil (FIAU; 100 nM, Movarek Biochemicals, Brea, CA, http://www.moravek.com) for positive and negative selection respectively. After 8 days, resistant clones were picked and seeded on a mouse embryonic fibroblast STO monolayer in individual wells of a 96 well plate to promote growth of low density ESC clones. Clones were screened for targeting events by PCR using standard methods. Targeted clones were detected and further characterized.

5.3 Southern Hybridization

Genomic DNA was isolated from RW4 and P-Olig2 ESCs using the ArchivePure DNA Cell/Tissue and Tissue Kit (5 Prime, Gaithersburg, MD, http://www.5prime.com). DNA-binding proteins were digested with Proteinase K. Genomic DNA was digested with 200 U HindIII or 50 U SpeI overnight at 37°C. Restriction enzymes were re-applied for an additional hour then DNA was ethanol precipitated. DNA restriction fragments were separated by electrophoresis. DNA was transferred to a Hybond-XL (GE Healthcare Biosciences, Piscataway, NJ, http://www.gehealthcare.com) membrane and crosslinked using a UV Stratalinker 2400 (Stratagene). DNA probes were prepared with the Rediprime Kit (GE Healthcare Bioscience) and [³²P] dCTP (Perkin-Elmer, Waltham, Massachusetts, http://www.perkinelmer.com). Probes were purified using illustra ProbeQuantG-50 columns (GE Healthcare Biosciences). Blots were hybridized in Rapid hybe (GE Healthcare Biosciences) for two hours at 65°C, then washed and visualized by autoradiography.

5.4 Cre-excision

To remove the floxed PGK-neo resistance cassette, 2×10⁶ ESCs were transfected with 5 μg of Cre recombinase expressing plasmid (p1411, gift of Tim Ley, Washington University). Removal of the PGK-neo cassette was confirmed using junction PCR. Cre-excised clones were further subcloned and re-validated by PCR and neomycin sensitivity.

5.5 Live Dead Assay

Live/Dead reagent (Invitrogen) consisting of calcien-AM and ethidium homodimer was used to visualize live and dead cells, respectively. Wells were washed with Dulbecco’s PBS and incubated with 1x Live/Dead reagent for 30 min at room temp. Fluorescent images were captured using MetaVue image analysis software (Molecular Devices, Sunnyvale, CA, http://www.moleculardevices.com) and a Nikon TE200S fluorescence microscope.

5.6 pMN Differentiation

For pMN induction, ESCs were exposed to RA (Sigma) and purmorphamine (EMD, Gibbstown, NJ, http://www.emdchemicals.com) in a 2⁻/4⁺ differentiation protocol. One million ESCs were aggregated into embryoid bodies (EBs) in 100-mm Petri dishes coated with a 0.1% agar solution in DFK5 media consisting of DMEM:F12 base media (Invitrogen) supplemented with 5% knockout serum replacement (Invitrogen), 50 μg/ml apo-transferrin (Sigma), 50 μM non-essential amino acids (Invitrogen), 5 μg/ml insulin (Sigma), 30 nM sodium selenite (Sigma), 100 μM β-mercaptoethanol, 5 μM thymidine, and 15 μM of the following nucleosides: adenosine, cytosine, guanosine, and uridine. EBs were allowed to form for 2 days in the absence of inducing factors, then split 1:5 and transferred to 6-well plates coated with 0.1% gelatin solution. Once plated, the EBs were grown for an additional 4 days in DFK5 supplemented with 2 μM RA and 250 nM – 1.5 μM purmorphamine. Media was changed every 2 days. In selected cultures, 2 ng/ml puromycin (Sigma) was added during the final 2 days of differentiation.

5.7 Quantitative Real Time Polymerase Chain Reaction (qRT-PCR)

The relative expression level of progenitor cell transcription factors and markers for MN differentiation were assessed using quantitative real-time polymerase chain reaction (qRT-PCR). EBs were lysed with Trizol reagent (Invitrogen) and RNA was isolated using an RNeasy kit (Qiagen, Germantown, MD, http://www.qiagen.com). Isolated RNA was used to synthesize cDNA for qRT-PCR analysis using the TaqMan 2-Step RNA-to-CT Mini Kit (Applied Biosystems, Carlsbad, CA, http://www.appliedbiosystems.com). TaqMan Gene Expression Assays (Table 1, Applied Biosystems), TaqMan Gene Expression Master Mix (Applied Biosystems) and cDNA were combined and qRT-PCR was performed using an Applied Biosystems 7000 thermocycler with the following PCR protocol: 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s, 55°C for 30 s and 72°C for 30 s. 6-carboxyfluorescein (FAM) fluorescent detection occurred during each 72°C cycle. Relative mRNA expression was reported as the number of cycles necessary for fluorescent intensity to increase exponentially, referred to as the threshold cycle (C_t). All target genes were normalized to ß-actin to account for differences in total mRNA content. Expression in EBs induced with 2 μM RA and 250 nM, 500 nM, or 1 μM purmorphamine was determined using the comparative ΔC_t method, with EBs receiving neither RA nor purmorphamine serving as the control group [30]. Results are reported as a fold difference in relative RNA expression over control EBs (n=3 for each condition).

Table 1.

List of Applied Biosystems Gene Expression Assay IDs

Gene	Primer ID
Dbx2	Mm01306497_m1
Irx3	Mm00500463_m1
Pax6	Mm00443072_m1
Olig2	Mm01210556_m1
Nkx2.2	Mm00839794_m1
Islet1	Mm00517585_m1
Hb9	Mm00658300_g1
Choline Acetyltransferase	Mm01221882_m1
β-Actin	Mm02619580_g1

5.8 Flow Cytometry

Differentiated ESC cultures were stained immediately following the 2⁻/4⁺ treatment protocol for flow cytometry analysis. Cultures were dissociated with trypsin-EDTA (0.25%; Invitrogen) for 15 min and triturated to form single cell suspensions. Excess volume of complete media was added to quench the trypsin, and cells were centrifuged for 5 min at 230×g. The media was aspirated and cells were fixed with 1% paraformaldehyde (Sigma). After fixation, the cells were permeabilized with 0.5% saponin (Sigma) solution for 20 min, and then blocked in 0.1% saponin solution containing 5% normal goat serum (NGS; Sigma). Cell suspensions were then incubated for 30 min in 0.1% saponin solution containing 2% NGS and one of the following primary antibodies: Nestin (Iowa Hybridoma Bank; 1:10), Isl1 (Iowa Hybridoma Bank; 1:50), Hb9 (MRN2; Iowa Hybridoma Bank; 1:25), and Olig2 (Millipore; 1:500). Cells were washed with PBS and appropriate Alexa Fluor secondary antibodies (1:200; Invitrogen) diluted in 0.5% saponin with 2% NGS were applied for 30 min. Finally cells were washed with PBS and incubated with Hoechst (1:1000; Invitrogen) for 5 min.

Stained cell suspensions were analyzed using a Canto II flow cytometer (Becton Dickinson, Franklin Lakes, NJ). For each group, 10,000 events were recorded. Subsequent analysis was performed using FloJo software (FloJo, Ashland, OR, http://www.flojo.com). Prior to population gating, debris was removed based on forward scatter versus side scatter and Hoechst fluorescence versus forward scatter plots. Flow cytometry control groups, consisting of cells stained with the secondary antibody only, were used to determine quadrant population gating parameters. Flow cytometry results are presented as the percentage of cells staining positive for each marker out of the total live cell population.

5.9 Immunocytochemistry

Cell distribution and identity was assessed in differentiated cultures using immunocytochemistry (ICC). Cell cultures were fixed with 1% paraformaldehyde for 30 min then permeabilized in 0.01% Triton X (Sigma) for 15 min. The cells were blocked with 5% NGS for 1 hour at 4°C and incubated overnight at 4°C in 2% NGS solution with one or more of the following primary antibodies: Oct 4 (1:500, Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com), Olig2 (1:500), Isl1 (1:20), Hb9 (1:20), Nestin (1:10), Neurofilament (NF, DSHB, 1:25), choline acetyl transferase (ChAT, Millipore, 1:400), oligodendrocytes (RIP, Millipore, 1:5000), and oligodendrocyte marker 4 (O4, Milllipore, 1:500). Primary antibody staining was followed by 3 washes in an excess volume of PBS for 15 min each. Each culture was then stained with the appropriate Alexa Fluor secondary antibodies (Invitrogen) for 1 hour at 4°C followed by an additional 3 washes in PBS. Cell nuclei were stained with the nuclei binding dye Hoechst (1:1000). Fluorescent images were captured using a MICROfire camera attached to an Olympus IX70 inverted microscope. Images were analyzed using Image Pro Express (Media Cybernetics, Silver Spring, MD, http://www.mediacy.com).

5.10 Undifferentiated ESCs

To quantify the occurrence of Oct4⁺ undifferentiated stems cells following the 2⁻ /4⁺ differentiation protocol, P-Olig2 cultures were fixed and stained with the Oct4 antibody. EBs containing at least one Oct4⁺ nuclei were counted and divided by the total number of EBs to determine the fraction of EBs containing undifferentiated stem cells. At least 50 EBs per sample were assessed for Oct4 expression.

5.11 Differentiation of pMNs

For pMN differentiation, 24-well plates were pre-coated with 0.01% poly(ornithine) solution (Sigma) then coated with 0.01 mg/mL laminin solution (Invitrogen) overnight. Induced 2⁻/4⁺ EBs were dissociated and plated at a density of 100,000 cells/ml in DFK5 media. After 5 days, media was replaced with Neurobasal media (Invitrogen) supplemented with 0.1% bovine serum albumin (Sigma) and 2% B27 (Invitrogen). Cells were allowed to differentiate an additional 9 days.

5.12 Statistical Analysis

For qRT-PCR and flow cytometry analyses, 3 replicates of each condition were performed. Statistical analysis was performed in Statistica software (version 5.5; StatSoft; OK, www.statsoft.com). Multiple comparisons statistics were accomplished using Scheffe’s post hoc test for analysis of variance (ANOVA) with a 95% confidence level. Values are reported as the mean plus or minus standard deviation.

Highlights.

• Greater progenitor motor neuron cell purity is needed to isolate motoneurons.

• One allele of the Olig2 gene was replaced with puromycin N-acetyl-transferase.

• Specific puromycin resistance in Olig2⁺ cells promotes survival during selection.

• Cells are enriched and differentiate into motoneurons and oligodendrocytes.

Abbreviations

pMNs

progenitor motor neurons

PAC

puromycin N-acetyl transferase (PAC)

NF

neurofilament

ChAT

choline acetyl transferase

MN

motoneuron

RA

retinoic acid

Shh

sonic hedgehog

Isl1

Islet1

GRE

gene regulatory elements

O4

oligodendrocyte marker 4

EB

embryoid body