Modeling sporadic juvenile ALS in iPSC-derived motor neurons explores the pathogenesis of FUS ^R503fs mutation

SUMMARY Human pluripotent stem cells (hPSCs) have opened new opportunities for understanding human development, modeling disease processes and developing new therapeutics. However, these applications are hindered by low-efficiency and heterogeneity of target cell types differentiated from hPSCs, such as motor neurons (MNs), as well as our inability to maintain the potency of lineage committed progenitors. Here, by using a combination of small molecules that regulate multiple signaling pathways, we develop a method to guide human embryonic stem cells to a near-pure population (>95%) of motor neuron progenitors (MNPs) in 12 days, and an enriched population (>90%) of functionally mature MNs in an additional 16 days. More importantly, the MNPs can be expanded for at least 5 passages so that a single MNP can be amplified to 1×104. This method is reproducible in human induced pluripotent stem cells and is applied to model MNdegenerative diseases and in proof-of-principle drug screening assays.

Amyotrophic lateral sclerosis (ALS) is a heterogeneous motor neuron disease for which no effective treatment is available, despite decades of research into SOD1-mutant familial ALS (FALS). The majority of ALS patients have no familial history, making the modeling of sporadic ALS (SALS) essential to the development of ALS therapeutics. However, as mutations underlying ALS pathogenesis have not yet been identified, it remains difficult to establish useful models of SALS. Using induced pluripotent stem cell (iPSC) technology to generate stem and differentiated cells retaining the patients' full genetic information, we have established a large number of in vitro cellular models of SALS. These models showed phenotypic differences in their pattern of neuronal degeneration, types of abnormal protein aggregates, cell death mechanisms, and onset and progression of these phenotypes in vitro among cases. We therefore developed a system for case clustering capable of subdividing these heterogeneous SALS models by their in vitro characteristics. We further evaluated multiple-phenotype rescue of these subclassified SALS models using agents selected from non-SOD1 FALS models, and identified ropinirole as a potential therapeutic candidate. Integration of the datasets acquired in this study permitted the visualization of molecular pathologies shared across a wide range of SALS models.