Advanced patient-specific microglia cell models for pre-clinical studies in Alzheimer’s disease

Background: Alzheimer’s disease (AD) is an incurable neurodegenerative disorder with a rapidly increasing prevalence worldwide. Current approaches targeting hallmark pathological features of AD have had no consistent clinical benefit. Neuroinflammation is a major contributor to neurodegeneration and hence, microglia, the brain’s resident immune cells, are an attractive target for potentially more effective therapeutic strategies. However, there is no current in vitro model system that captures AD patient-specific microglial characteristics using physiologically relevant and experimentally flexible culture conditions. Methods: To address this shortcoming, we developed novel 3D Matrigel-based monocyte-derived microglia-like cell (MDMi) mono-cultures and co-cultures with neuro-glial cells (ReNcell VM). We used single-cell RNA sequencing (scRNAseq) analysis to compare the transcriptomic signatures of MDMi between model systems (2D, 3D and 3D co-culture) and against published human microglia datasets. To demonstrate the potential of MDMi for use in personalized preclinical strategies, we generated and characterized MDMi models from sixteen AD patients and matched healthy controls, and profiled cytokine responses upon treatment with anti-inflammatory drugs (dasatinib and spiperone). Results: MDMi in 3D exhibited a more branched morphology and longer survival in culture compared to 2D. scRNAseq uncovered distinct MDMi subpopulations that exhibit higher functional heterogeneity and best resemble human microglia in 3D co-culture. AD MDMi in 3D co-culture showed altered cell-to-cell interactions, growth factor and cytokine secretion profiles and responses to amyloid-β. Drug testing assays revealed patient- and model-specific cytokine responses. Conclusion: Our study presents a novel, physiologically relevant and AD patient-specific 3D microglia cell model that opens avenues towards improving personalized drug development strategies in AD. We characterized monocyte-derived microglia (MDMi) cultured in 2D, 3D, and 3D co-cultured with ReNcell VM, with our data demonstrating that 3D co-culture closely resembles human brain microglia compared to 3D MDMi and 2D MDMi. The experimental design involved the collection of longitudinal blood samples from a single healthy male control donor under 40 years of age, free of dementia, and without a family history of Alzheimer's disease (AD). Blood collections were performed bi-monthly over a four-month period, resulting in multiple time points. Using EDTA tubes, peripheral blood mononuclear cells (PBMC) were isolated following the SepMate50 protocol. MDMi were cultured for 14 days (2D) or 35 days (3D) under a culture medium containing IL-34 and GM-CSF. 3D mono-culture and 3D co-culture were grown in thick Matrigel, while 2D was cultured on Matrigel-coated plates. For RNA collection, MDMi from all conditions were isolated and sorted for CD11b+ (thereby excluding leukocytes and ReNcell VM) and fixed according to the Chromium fixed RNA profiling workflow. Library construction involved simultaneous processing of all samples.