All-trans-retinoid acid induces the differentiation of encapsulated mouse embryonic stem cells into GABAergic neurons

Abstract

Embryonic stem (ES) cells are pluripotent cells that can differentiate into all three main germ layers: endoderm, mesoderm, and ectoderm. Although a number of methods have been developed to differentiate ES cells into neuronal phenotypes such as sensory and motor neurons, the efficient generation of GABAergic interneurons from ES cells still presents an ongoing challenge. Because the main output of inhibitory GABAergic interneurons is the gamma-aminobutyric-acid (GABA), a neurotransmitter whose controlled homeostasis is required for normal brain function, the efficient generation in culture of functional interneurons may have future implications on the treatment of neurological disorders such as epilepsy, autism, and schizophrenia. The goal of this work was to examine the generation of GABAergic neurons from mouse ES cells by comparing an embryoid body-based methodology versus a hydrogel-based encapsulation protocol that involves the use of all-trans-retinoid acid (RA). We observed that (1) there was a 2-fold increase in neuronal differentiation in encapsulated versus non-encapsulated cells and (2) there was an increase in the specificity for interneuronal differentiation in encapsulated cells, as assessed by mRNA expression and electrophysiology approaches. Furthermore, our results indicate that most of the neurons obtained from encapsulated mouse ES cells are GABA-positive (∼87%). Thus, these results suggest that combining encapsulation of ES cells and RA treatment provide a more efficient and scalable differentiation strategy for the generation in culture of functional GABAergic interneurons. This technology may have implications for future cell replacement therapies and the treatment of CNS disorders.

Figure 1. Cell encapsulation increases the efficiency of ES neuronal differentiation and improves ES cell viability

A) Quantitative analysis of immunofluorescence experiments showing the percentage of neuronal differentiation in encapsulated (Enc) versus non-encapsulated (NE) ES cells. Cells treated with vehicle only (0RA) or 5μM RA (5RA) were examined at day 8 for the expression of β-tubulin III by immunofluorescence. The percentage of cells expressing β-tubulin III was calculated based on the total number of cells as determined by nuclear staining with DAPI. Quantitative analyses from triplicate experiments demonstrated that there was a 2-fold increase in β-tubulin expression in encapsulated versus non-encapsulated cells after RA treatment. B) Cell viability was determined by the trypan blue exclusion method. These results demonstrated that encapsulation significantly increased the viability of mouse ES cells as compared to non-encapsulated cells. *p<0.05 by one-way ANOVA. Means ± SEM are shown.

Figure 2. Encapsulation enhances the generation of neuronal precursors in RA-treated cells

Encapsulated ES cells or EBs were treated with 5 μM RA and harvested at day 8. Encapsulated aggregates and EBs were either sectioned or disaggregated for the determination of Nestin expression by immunofluorescence. A) Representative sections of encapsulated (ENC) or non-encapsulated (NE) cells showing Nestin expression. B) Statistical analysis showing the percentage of Nestin-positive ENC or NE cells in disaggregated samples. The total number of cells was determined by nuclear staining with DAPI. These experiments demonstrated that encapsulation increased the percentage of neuronal precursors by ~1.6-fold. Scale bar, 10 μm. *p<0.05.

Figure 3. Characterization of day 8 encapsulated versus non-encapsulated ES cells by RT-PCR

The mRNA expression levels of various neuronal and non-neuronal markers was examined by non-quantitative RT-PCR in day 8 Encapsulated (Enc) or Non-encapsulated (NE) cells treated with vehicle-only (0RA) or 5 μM RA (5RA). Total RNA was reverse transcribed and cDNA was used as template for PCR employing specific primer pairs for the following markers: Rex1 and Nanog (stem cell markers), GFAP (differentiated astrocytes), TH (dopaminergic neurons), GLT1 (astroglial and glutamatergic neurons), GAD1 (also called GAD67; GABAergic neurons), and 36B4 (a ubiquitously-expressed gene used as a loading control). -RT: no reverse transcriptase. B= Adult mouse brain tissue. ES = Embryonic stem cells.

Figure 4. Neurons Generated from Encapsulated mouse ES Cells are GABA-positive

Neurons obtained at the 4M stage were fixed and the expression of GABAergic markers was examined by immunofluorescence. A) Co-expression of the GABAergic markers GAD65/67 (GAD) and GABA in differentiated E1 ES cells treated with vehicle-only (0RA) or 5 μM RA (5RA). Encapsulated E1 cells were harvested at day 8 followed by plating on PDL/laminin-coated dishes. The expression of GAD and GABA was examined by immunofluorescence at 4M. B) Quantitative analysis from independent triplicate experiments showing the percentage of GABA-expressing cells based on the total number of cells as determined by nuclear DAPI staining. C) GABAergic marker expression in differentiated J1 cells at 4M. D) The percentage of GABA-expressing cells was determined from triplicate experiments. The results obtained using E1 and J1 ES cells were similar. **p<0.001 by one-way ANOVA. Means + SEM are shown.

Figure 5. Neurochemical characterization of differentiated GABAergic neurons

Neurons obtained from mouse E1 ES cells were examined at 4M for the expression of A) somatostatin (SM) and B) parvalbumin (PV) using immunofluorescence. Both SM- and PV-expressing GABAergic subtypes were identified among the differentiated neurons. Co-staining with β-tubulin III demonstrates the neuronal identity of cells expressing SM or PV. C) The percentage of β-tubulin III-positive cells expressing SM or PV was determined from triplicate experiments.

Figure 6. Whole-cell patch-clamp recordings in differentiated embryonic stem cells

A) Average outward currents recorded from neuronal differentiated cells and undifferentiated controls for 60 s. Currents were extracted at +50 mV from a voltage ramp ranging from −100 mV to +100 mV and −80 mV holding potential (n=4 cells/group; mean+SEM). B) The current-voltage relationship (I/V) is typical for voltage-gated delayed rectifier K⁺ channels in the neuronal differentiated cell, but was absent in control undifferentiated cells. C) Average peak currents from panel A at 60 s after establishment of whole-cell configuration (*P<0.05). D) Data extracted at the respective voltage from the differentiated neuronal cell in panel B.

Figure 7. Effect of depolarization on Voltage-Gated Delayed Rectifier K⁺ currents

A) Current-voltage relationship (I/V) from a differentiated neuronal cell extracted at +50mV after establishment of whole-cell configuration with a holding potential of 0mV. Note the inhibitory effect of voltage on channel currents. B) Outward current from panel A showing the rapid channel closure after establishment of whole-cell configuration.