Directed differentiation of Human induced pluripotent stem cells generates active motor Neurons

Introduction:

iPS cells may provide a means of generating functional motor neurons for regenerative medicine. Many diseases, such as amyotrophic lateral sclerosis and spinal muscular atrophy, result in the loss of motor neurons. Previous studies have shown the ability of human embryonic stem cells to generate motor neurons. Hence, these methods, when applied to iPS cells, yield motor neurons that are electrophysiologically functional and seem to follow a normal development. This study suggests that iPS cells could be used in regenerative medicine to repair the functional loss of motor neurons and could also provide a means of modeling the developmental progression of motor neuron diseases in vitro.

Methods:

Control: A well studied hESC line – HSF1

Two different protocols were used to generate functional mature motor neurons from iPS cells.

Embryoid body differentiation protocol used EBs cultured on an hESC media lacking FGF2 for 1 week and were treated with retinoic acid for one week. The EBs were also treated with a Sonic Hedgehog pathway agonist. This resulted in the neuralization of the embryoid bodies, indicated by the expression of neural progenitor markers such as Brn2, Sox3, and Pax6. Progressive differentiation of the pluripotent cells to neural progenitors was observed, though only a certain fraction of the HSF1 and iPS cells showed neural markers detected via immunostaining.

An adherent approach was used to generate neural rosettes from HSF1 and multiple iPS cell lines (iPS1, iPS2 and iPS18). The rosettes were mechanically isolated and plated on to laminin coated dishes along with Retinoic acid and Shh for 1 week. Neurotrophic factors such as BDNF, CNTF and GDNF were added and Shh concentration was lowered by 75% to allow the cells to differentiate for 3-5 weeks. Transition of the cells from being nestin-positive neural progenitors to mature motor neurons was detected via immunostaining.

The differentiation protocols were used to show that once specified to a neuronal fate, the hESCs and the iPS cells were comparable in generating motor neurons.

Antibody immunostaining was performed to detect neural progenitor markers and fluorescence and DIC images were collected. Whole cell, patch clamp analysis was also performed for electrophysiological activity of the differentiated neurons.

Results:

With regards to the differentiation efficiencies of the HSF1 cell like and the iPS cell line, the authors reported that there were significant differences between the two. They suggested that the discrepancy was due to the variability between different cell lines, which might have attributed to the inconsistency in efficiency. However, the authors went on to say that among those EBs who showed a preference to neural fate (with the expression of Brn2, Sox3, and Pax6), the number of differentiated cells were similar (Fig. 1). Upon further treatment of the cells with retinoic acid, Shh agonists and neurotrophic factors, the EBs were cryosectioned and stained for motor neuron progenitor markers Nkx6.1 and Olig2 (Fig 2).

Figure 1: hESC cells and iPS cells derived from EB protocol stained for neural progenitor markers Brn2, Sox3 and Pax6.

Figure 2: HSF1 and hiPS2 cells stained for neural progenitor markers as well as mature motor neuron markers (Nxk6 and Olig2). Percentages for Sox3+ cells: ~59.1% HSF1and ~57.6% of hiPS2 cells.

Cells obtained from the adherant approach were stained for neural progenitor markers (nestin) as well as for mature motor neuron markers (BIII-tubulin, chat and Islet1+). Similar percentages of Islet1+ cells were reported for the HSF1 and hiPS2 cells were reported, indicating that the differentiation efficiency was similar between the two cell lines. These cells were also stained for a Hb9 activity (a transcription factor expressed in mature motor neutons). The authors used Hb9-driven green fluorescent protein reporter gene to identify activity. Hoxa5 gene staining shows that the cells expressed a rostral cervical character. (Fig.3)

Figure 3: HSF1 cells and hiPS2 cells were stained for neural progenitor markers (nestin) as well as for mature motor neuron markers (BIII-tubulin, chat and Islet1+) (3A-H). HSF1, hiPS1 and hiPS18 derived motor neurons were stained for ChAT activity and Hoxa5 as markers for spinal cord neurons.

It is a characteristic of motor neurons to fire repetitive action potentials as a response to current injection. Whole cell patch clamping was used to asses the excitability of the neurons. Roughly half of the Hb9:GFP+ neurons (derived from iPScs and hESCs) responded with repetitive action potentials. The cells also showed choline acetyltransferase activity which is typically seen in motor neurons as a response to electrical stimulation. (Fig 4). These results demonstrate the potential of iPS cells to generate active motor neurons.

Figure 4: Electrophysiological properties of HSF1 and hiPS2 cells that stained positive for ChAT and expressed Hb9-GFP reporter gene.

Critique:

Though this paper shows the capacity of iPS cells to become motor neurons, there are several inconsistencies in their results. Multiple iPSC clones were used for these experiments, but the date often showed only one cell line or another. They did not mention any data on the behavior of physiological neurons as a positive control, despite recurrently placing emphasis on the use of iPSC-derived motor neurons for regenerative medicine. Most of this study was exclusively qualitative, without any focus on the extent of gene expression in the derived neurons. Additionally, the electrophysiological studies did not use the same amount of current to test motor neurons derived from different cell lines. The frequency of action potentials also varied between cell lines. Further quantitative experiments and better controls should be done to better understand the behavior of iPSC derived motor neurons.