This is a summary of Klimmt et al. A reproducible three-dimensional model of human brain tissue to investigate physiological and disease-associated microglia phenotypes. Published in Nature Neuroscience (2026). DOI: 10.1038/s41593-026-02367-0

The challenge

Human brain tissue models are essential for studying neurodevelopment and neurodegenerative diseases, but current systems face major limitations. Neural organoids and 2D co-cultures often lack reproducibility, show immature cell states, and fail to support long-term maintenance of homeostatic microglia. This hampers investigations into microglial contributions to Alzheimer’s disease and other brain disorders, and limits the physiological relevance of drug testing in fully human systems.

Our approach

We developed a three-dimensional cortical brain tissue model (3BTM) by aggregating iPSC-derived neurons, astrocytes and microglia. Using multi-line validation, proteomics, single-cell RNA-seq, live imaging and functional assays, we characterized maturation, ECM formation and microglial phenotypes. We then engineered 3BTMs with synergistic APP mutations to model Alzheimer’s disease and tested anti-amyloid immunotherapy to assess therapeutic responses.

Our findings

3BTMs showed high reproducibility, stable size, and long-term viability without necrotic cores. Neurons, astrocytes and microglia matured toward in vivo-like states, with brain-specific ECM deposition and synapse formation. Microglia survived for over six months, adopted ramified morphologies, expressed homeostatic markers, and responded dynamically to injury. AD-modeled 3BTMs developed amyloid deposition, increased phospho-tau and neuroinflammation, with microglia shifting to disease-associated transcriptional states. Anti-amyloid immunotherapy cleared deposits and reversed disease signatures in glia.