These organoid technologies have been pioneered for the smal

These organoid technologies have been pioneered for the small intestine (Sato et al., 2009) and later extended also to other organs or parts of organs (Nakano et al., 2012). Recently, protocols for the generation of human brain-like organoids have been developed, including protocols for cerebral (Lancaster et al., 2013), cerebellar (Muguruma et al., 2015), midbrain-like (Jo et al., 2016), and forebrain organoids (Qian et al., 2016). These organoids provide a proof of concept that human iPSCs can indeed differentiate into various cell types and even self-organize with a specific spatial orientation, which recapitulates key features of the human brain. These pde inhibitor organoids have been used successfully to model a genetic form of microcephaly (Lancaster et al., 2013) as well as Zika virus-induced microcephaly (Qian et al., 2016). Importantly, thus far all brain organoids have been generated directly from iPSCs; however, evidence suggesting that these organoids could also be derived from more fate-restricted neural stem cells is lacking. The utilization of neural stem cells as the starting population has the advantage that already patterned cells might differentiate into the desired structures more efficiently (cheaper, faster cell doublings, ease of handling, and so forth). Furthermore, while the generation of certain brain structures, such as the cerebral cortex or cerebellum, is meanwhile well described, a higher degree of prepatterning seems to be required for the generation of other highly specialized structures. This is particularly true for brain regions severely affected by neurodegenerative disorders, such as Alzheimer\'s disease (hippocampus) or Parkinson\'s disease (PD; substantia nigra). To address these challenges, we used our previously described human neuroepithelial stem cell (NESC) culture system (Reinhardt et al., 2013) and differentiated these NESCs under dynamic conditions into human midbrain-specific organoids.
Results
Discussion The presence of astrocytes is crucial for the formation of synapses and regular neuronal activity (Chung et al., 2015). Astrocytes are specified later than neurons during development (Chaboub and Deneen, 2013; Molofsky et al., 2012). Accordingly, hMOs show robust astrocyte immunoreactivity only after 61 days of differentiation. Furthermore, synaptic connections, consisting of a direct contact between pre- and postsynapses, are detectable in the hMOs. These synaptic contacts are the prerequisite for electrophysiological and neuronal network functionality, which we indeed detected in the hMOs by Ca2+ imaging and MEA measurements. In addition, fast information transmission between neurons depends on axonal myelination, which is achieved by oligodendrocytes. In most stem cell-based differentiation protocols, the differentiation into oligodendrocytes is extremely inefficient (Bunk et al., 2016; Jablonska et al., 2010). However, in the present approach we achieved a robust differentiation into oligodendrocytes and a high degree of neurite myelination. Neurites in these hMOs are ensheathed by oligodendrocytes and even structures such as the nodes of Ranvier, which are of critical importance for saltatory transmission of signals in axons (Faivre-Sarrailh and Devaux, 2013), become apparent. Compared with other pioneering human brain organoid systems, we are able to generate organoids of a remarkable size (up to 2 mm in diameter) with high reproducibility. Importantly, it was possible to reproducibly generate hMOs based on iPSC lines that come from different origins such as cord blood and fibroblasts. Additionally these fibroblasts have been sampled from individuals at ages from 53 years to 81 years (Table S1). Importantly, in contrast to all other human brain organoid systems, our starting population of cells are not iPSCs but NESCs (Reinhardt et al., 2013), which allows us to achieve an efficient directed differentiation. Our approach is fully focused on the midbrain, as shown by the abundant presence of neurons with mDN identity. These neurons are asymmetrically distributed in a discrete cluster. This asymmetry mirrors a unique feature of the human brain, where the soma of mDNs reside in the substantia nigra. Furthermore, the presence of neuromelanin, which is a unique feature of the primate brain, is an interesting finding.