Modeling striatal development and disease with human pluripotent stem cells.

The striatum plays a central role in motor control, cognition, reward processing, and habit formation, and its dysfunction is implicated in a broad spectrum of neurological and psychiatric disorders. Although animal models have provided important insights into striatal development and disease mechanisms, species-specific differences in cellular composition, developmental timing, and circuit organization limit their translational relevance to the human brain. In this context, human pluripotent stem cells (PSCs), including embryonic stem cells and induced pluripotent stem cells, have emerged as valuable platforms for modeling human striatal development and pathology in vitro. In this review, we summarize current approaches for generating striatal cell types from PSCs, with a particular focus on medium spiny neurons (MSNs), the principal projection neurons of the striatum. We discuss key developmental principles underlying dorsal and ventral striatal specification and highlight the protracted maturation of human MSNs, which may contribute to human-specific disease vulnerability. Advances in differentiation strategies, including small molecule-based patterning, transcription factor-driven induction, and three-dimensional organoid and assembloid systems, have progressively improved the efficiency, reproducibility, and cellular complexity of PSC-derived striatal models. We further review applications of PSC-derived striatal systems in disease modeling, noting that most studies to date have focused on Huntington's disease, where these models have revealed early developmental, transcriptional, synaptic, and network-level abnormalities. More recent studies have begun to extend these approaches to other neurological conditions and to incorporate circuit-level analyses using cortico-striatal assembloids. In parallel, the growing availability of single-cell and single-nucleus transcriptomic datasets from the human striatum provides powerful reference frameworks for benchmarking the identity and maturation state of PSC-derived striatal cells. Finally, we discuss current challenges and limitations of PSC-based striatal models, including incomplete maturation, limited representation of non-neuronal cell types, and restricted applicability to psychiatric disorders. We propose that continued integration of developmental biology, public multi-omics resources, and advanced in vitro modeling strategies will be essential for advancing human striatal models and expanding their utility in translational neuroscience.