Astrocytes from human pluripotent stem cells: Ontogenesis, disease modeling, and therapeutic discovery.

摘要

Astrocytes are known to play major roles in normal brain development and pathogenesis of neurological diseases, yet a human specific model system has been lacking for studying ontogeny and functions in normal and diseased states. Here, I have developed a novel chemically defined system to differentiate human embryonic stem cells (hESCs) and induced pluripotent stem cells (hiPSCs) to a uniform population of functional astrocytes. I found that the differentiation process follows developmental principles similar to that of embryonic human brain development. I further discovered that regionally and functionally specialized astrocyte subtypes can be efficiently generated by patterning the early neuroepithelial cells with specific sets of morphogens, strongly suggesting that astrocyte subtypes are specified during early development. These in vitro generated human astrocytes exhibit astrocyte-specific voltage gated- and glutamate stimulated-membrane conductance, propagation of calcium waves in response to stimuli, and formation of endfeet around blood vessels following transplantation into the nervous system. They thus offer a useful source of bona fide human astrocytes for drug discovery and potential cell therapy.;Astrocytes are implicated in the disease propagation of amyotrophic lateral sclerosis (ALS). Therefore, I have generated transgenic hESC-derived spinal astrocytes that express disease forms of copper-zinc superoxide dismutase (SOD1), e.g. G85R and A4V mutations, as well as wild type SOD1 as a novel ALS model system. Human astrocytes expressing mutant SOD1, but not controls, displayed disease related phenotypes, including accumulation of large protein aggregates, mitochondrial swelling, altered calcium wave propagation distance and ATP release, and toxicity towards spinal motoneurons. Mass spectrometry based proteomic analysis of the mutant SOD1-expressing human astrocytes, compared to controls, revealed changes in secreted protein abundance related to the function of multiple inflammation-related proteins and ALS biomarkers, all potential targets for therapeutic strategies.;In addition, by differentiating astrocytes from GFAP mutant-specific hiPSCs derived from an Alexander disease patient, I discovered a disease-specific phenotype that may be targeted for cell based drug screening. Taken together, the system I developed allows generation of large quantities of functionally normal or diseased astrocytes, providing a novel human cell source for developmental analysis, disease modeling, and therapeutic applications.