Cerebral organoids are tractable models of early brain development. Reproduced with permission from Li et al., 2017.

The human brain is unusually large and densely innervated, mainly because of an expanded cerebral cortex, the region associated with consciousness and higher brain function. The processes that give rise to this important structure and the diseases that affect it are of great interest to scientists like Arnold Kriegstein at the University of California, San Francisco and Rudolf Jaenisch at the Massachusetts Institute of Technology. In new work, both use cerebral organoids as a way to access early stages of brain development and to model malformations.

Organoids are stem-cell-derived, three-dimensional cultures that self organize to some extent and include multiple cell types and features of a particular organ. Kriegstein, with Marina Bershteyn, Tomas Nowakowski and their colleagues, generated cerebral organoids using induced pluripotent stem cells from individuals with Miller-Dieker syndrome (MDS), a genetic disease characterized by a 'smooth' or lissencephalic brain that lacks cortical folds (Bershteyn et al., 2017). They used the pioneering protocol of Yoshiki Sasai, in which dissociated stem cells reaggregate in suspension and then differentiate into cortical organoids in the presence of growth factor cues.

Their work successfully recapitulated key features and stages of lissencephaly, including a combination of cell death and division defects in neuroepithelial progenitors, as well as cell-autonomous neuronal migration defects in MDS compared to wild type organoids. Cerebral organoids do have limitations, however. In particular, they currently only model fetal development in the first two trimesters, making them less suited for neurodegeneration and other diseases with later onset. “I think it's premature to look at anything that relates to a cortical circuit in an organoid,” adds Kriegstein. And missing cell types, including immune cells, can make it tricky to interpret disease phenotypes.

Jaenisch and his team were driven by the question of why rodent brains are smooth whereas primate brains are folded (Li et al., 2017). They used a robust protocol, developed by collaborators Madeline Lancaster and Juergen Knoblich, which also starts with free-floating stem cell aggregates that are then cultured in extracellular matrix. In matrix, multiple brain regions can spontaneously develop within a single organoid. The team discovered that genetic lesions in the anti-proliferative gene PTEN cause neural progenitor populations to expand and delay neuronal differentiation, but only lead to folding in human and not similarly cultured rodent cortical organoids. The exact mechanism of folding in human embryos is still somewhat unclear, however, as it occurs later and involves a different cortical structure.

Jaenisch also sees directions for improving organoids. For example, neurons begin to die once organoids reach a certain size. “One would probably like to bioengineer something better,” he says, referring to a circulation system that would allow larger and longer-lived organoids. Variability is also an issue; for now, large numbers of organoids must be analyzed until protocols produce more consistent results.

Despite some limitations, organoids provide powerful and tractable models of early brain development and disease. The Jaenisch team found that infection with Zika virus, which has been associated with microcephaly, disturbed cortical growth and folding, and both groups are developing further studies of viral infection using organoids.