Rapid, efficient, and simple motor neuron differentiation from human pluripotent stem cells

Abstract

Background: Human pluripotent stem cells (hPSCs) are being applied in regenerative medicine and for the in vitro modeling of human intractable disorders. In particular, neural cells derived from disease-specific human induced pluripotent stem cells (hiPSCs) established from patients with neurological disorders have been used as in vitro disease models to recapitulate in vivo pathogenesis because neural cells cannot be usually obtained from patients themselves.

Results: In this study, we established a rapid, efficient, and simple method for efficiently deriving motor neurons from hPSCs that is useful for pathophysiological analysis and the development of drugs to treat motor neuron diseases. Treatment with GSK3β inhibitors during the initial phase of differentiation in combination with dual SMAD inhibition was sufficient to induce PAX6 (+) and SOX1 (+) neural progenitors within 1 week, and subsequent treatment with retinoic acid (RA) and purmorphamine, which activates sonic hedgehog (SHH) signaling, resulted in the highly efficient induction of HB9(+) and ISL-1(+) motor neurons within 2 weeks. After 4 weeks of monolayer differentiation in motor neuron maturation medium, hPSC-derived motor neurons were shown to mature, displaying larger somas and clearer staining for the mature motor neuron marker choline acetyltransferase (ChAT). Moreover, hPSC-derived motor neurons were able to form neuromuscular junctions with human myotubes in vitro and induced acetylcholine receptor (AChR) clustering, as detected by Alexa 555-conjugated α-Bungarotoxin (α-BTX), suggesting that these hPSC-derived motor neurons formed functional contacts with skeletal muscles. This differentiation system is simple and is reproducible in several hiPSC clones, thereby minimizing clonal variation among hPSC clones. We also established a system for visualizing motor neurons with a lentiviral reporter for HB9 (HB9 (e438) ::Venus). The specificity of this reporter was confirmed through immunocytochemistry and quantitative RT-PCR analysis of high-positive fractions obtained via fluorescence-activated cell sorting (FACS), suggesting its applicability for motor neuron-specific analysis.

Conclusions: Our motor neuron differentiation system and lentivirus-based reporter system for motor neurons facilitate the analysis of disease-specific hiPSCs for motor neuron diseases.

Fig. 1

Rapid, efficient and simple motor neuron differentiation from human pluripotent stem cells. a Schematic presentation of motor neuron differentiation from hPSCs. b Time-course analysis of the expression of PAX6, SOX1, and NGN2 in EBs via quantitative RT-PCR analysis. Control, 0.09 % DMSO. DS, dorsomorphin and SB431542. DSB, dorsomorphin, SB431542 and BIO. n = 4, mean ± SEM. *, p < 0.05 (Student’s t test). c Time-course analysis of the expression of PAX6, SOX1, NGN2, OLIG2, NKX2.2, HB9, and ISL-1 mRNA in EBs and during monolayer motor neuron (MN) differentiation. n = 4, mean ± SEM. d Immunocytochemical analysis of HB9, ISL-1, and βIII-Tubulin 1 week after monolayer differentiation. Scale bar, 100 μm. e Quantitative analysis of the number of the HB9⁺ and ISL-1⁺ cells. n = 3, mean ± SEM

Fig. 2

Long-term culture of hESC-derived motor neurons. a Immunocytochemical analysis of HB9, ISL-1, βIII-Tubulin, and ChAT after 1, 2, or 4 weeks of monolayer differentiation. Scale bar, 100 μm, and 20 μm (inset). b Quantification of the number of the cells positive for each marker. n = 4, mean ± SEM. *, p < 0.05 (Student’s t test). c Time-course analysis of the expression of PAX6, SOX1, NGN2, OLIG2, NKX2.2, HB9, ISL-1, and ChAT during monolayer differentiation via quantitative RT-PCR. n = 4, mean ± SEM. d Western blot analysis of the expression of the HB9, ISL-1, and ChAT proteins during monolayer differentiation. e Quantitative analysis of the expression of the HB9, ISL-1, and ChAT proteins through densitometry using ImageJ. The protein expression levels are normalized to β-Actin. n = 4, mean ± SEM. *, p < 0.05, (Student’s t test)

Fig. 3

Co-culture of hESC-derived motor neurons with myotubes. a Motor neurons derived from KhES1 cells were plated on myotubes derived from the human myoblast cell line Hu5/E18 to form end plate-like structures. Scale bar, 20 μm. b Immunocytochemical analysis of βIII-Tubulin and MyHC after 3 days of co-culture with motor neurons and myotubes. AChR clusters were visualized with Alexa 555-conjugated α-Bungarotoxin (α-BTX, white allows). Scale bar, 50 μm

Fig. 4

Derivation of motor neurons from hiPSCs. a Immunocytochemical analysis of motor neurons derived from 201B7, TIGE-9 and YFE-16 for HB9, ISL-1, and βIII-Tubulin after 2 weeks of monolayer differentiation and ChAT after 4 weeks of monolayer differentiation. Scale bar, 100 μm. b Quantitative analysis of the number of the cells positive for motor neuron markers in cultures derived from each hiPSC clone. n = 3, mean ± SEM. c Time-course analysis of the expression of HB9, ISL-1, and ChAT during monolayer differentiation of hiPSC-derived motor neurons via quantitative RT-PCR. n = 3, mean ± SEM. d Western blot analysis of the expression of the HB9, ISL-1, and ChAT proteins during monolayer differentiation of hiPSC-derived motor neurons. B7, 201B7. T9, TIGE-9. Y16, YFE-16. e Quantitative analysis of the expression of the HB9, ISL-1, and ChAT proteins through densitometry using ImageJ. Protein expression levels are normalized to β-Actin. n = 3, mean ± SEM. *, p < 0.05, (Student’s t test)

Fig. 5

Replacement of small molecule compounds with more specific and less toxic inhibitors. a Morphology of EBs, which were derived from 201B7, treated with vehicle (DMSO), DS, DSB, or DSC on day 4. Massive cell death was observed after the treatment of DSB. Scale bar = 200 μm. Control, 0.09 % DMSO. DS, dorsomorphin and SB431542. DSB, dorsomorphin, SB431542, and BIO. DSC, dorsomorphin, SB431542, and CHIR99021. b The number of cells harvested from DSB- and DSC-treated EBs on day 14 per 100-mm culture dish of starting hiPSCs. n = 4, mean ± SEM. *, p < 0.05 (Student’s t test). c The number of cells harvested from DSC- and LSC-treated EBs on day 14 per 100-mm culture dish of starting hiPSCs. n = 4, mean ± SEM. LSC, LDN193189, SB431542, and CHIR99021. d Immunocytochemical analysis of motor neurons derived from DSC- or LSC-treated EBs for HB9, ISL-1 and βIII-Tubulin on day 7 of monolayer differentiation. Scale bar, 100 μm. d Quantitative analysis of the number of the cells positive for HB9 and ISL-1. n = 4, mean ± SEM. e Time-course analysis of the expression of PAX6, SOX1, NGN2, OLIG2, NKX2.2, HB9, and ISL-1 in EBs and during monolayer differentiation of motor neurons (MN) via quantitative RT-PCR. n = 4, mean ± SEM

Fig. 6

Visualization of motor neurons using the HB9 ^e438::Venus reporter lentivirus. a Live-cell imaging of HB9 ^e438::Venus⁺ motor neurons 3 days after lentivirus infection. Scale bar, 100 µm. b Immunocytochemical analysis of motor neurons for HB9, ISL-1, ChAT, and Venus (GFP antibody) at 1, 2, and 4 weeks after HB9 ^e438::Venus lentivirus infection. Scale bar, 100 μm. c Quantitative analysis of the number of cells positive for HB9, ISL-1, and ChAT among Venus⁺ cells. n = 4, mean ± SEM.*, p < 0.05 (Student’s t test) d Histograms of HB9 ⁴⁸³::Venus lentivirus-infected cells (Green), background lentivirus (β-glo::Venus)-infected cells (Yellow), and uninfected control cells (Gray) via flow cytometry. The HB9 ^e438::Venus⁺ cells were divided into 4 fractions: a negative fraction (Neg), in which the fluorescence intensities were equivalent to an uninfected negative control, and a low-positive fraction (Low), a middle-positive fraction (Middle), and a high-positive fraction (High), in which the fluorescence intensities were equivalent to the lowest 1/3, the next lowest 1/3, and the highest 1/3 of fluorescence intensities within the positive fraction, respectively. e The expression of Venus, ISL-1, HB9, and ChAT in each fraction determined via quantitative RT-PCR. n = 3, mean ± SEM. A significant increase in the expression of Venus, HB9, and ISL-1 in the high-positive fraction was observed (p < 0.05, Kruskal-Wallis test)