Human iPSC model of Cerebro-Vascular Interactions in ADRD - We propose to generate and validate a novel model that incorporates human cerebral cortical neurons and glia, vascular endothelial cells, neural crest pericytes and microglia from human induced pluripotent stem cells (iPSCs). The goal is a model with improved construct, face, and predictive validity to study the cerebro-vascular interactions common to Alzheimer’s Disease-Related Dementias (ADRD). The NINDS 2022 ADRD summit report describes the need for such a model. In prior studies we developed a reproducible human 3D cerebral cortex organoid model tested across 63 iPSC lines with 90% success. It improved efficiency in organoid production from 10% to approaching 100%. Organoids grown with the protocol have uniform size, shape, predictable growth curves and cell composition. Single cell transcriptome studies demonstrate high overlap between iPSC lines from different donors, an important advance for studies of both genetic and sporadic disease. We also recently developed an iPSC-derived 3D vasculature model forming robust, vessel-like structures of endothelial cells and pericytes in an inert hydrogel. Preliminary studies demonstrate that these two advanced models can be integrated to generate a human cell 3D cerebro-vascular model. In Aim 1, the R61 phase, we will optimize the incorporation of these two models to achieve greatest reproducibility of cell composition reflecting vascularized cerebral cortical tissue in vivo. The model will be generated from at least three different PSEN1 mutation iPSC lines versus healthy controls. There is significant need for this model given that current PSEN1 mutation transgenic and knock-in mouse models lack several key features of neuro- and vascular pathology seen in human disease. Quantitative metrics will be used to assess efficiency and reproducibility of the model and its ability to reproduce PSEN1 mutation cortical pathology, including cerebral amyloid angiopathy (CAA). With the achievement of prospectively determined quantitative criteria, we will advance to the R33 phase of the project. In Aim 2, the PSEN1 mutation models will be examined for face, construct, and predictive validity. This phase will include collaborators with expertise in examining PSEN1 mutation carrier tissue pathology, brain single nuclear transcriptomes, metabolomics, lipidomics, and cerebrospinal fluid (CSF) composition. We will assess established characteristics of PSEN1 mutation-induced cerebral pathology, such as altered expression of A40,42 and 43, and tau. Validation tests will include those used for clinical evaluation where test sensitivity, dynamic range and predictive value are known. The predictive validity of the model will be assessed by comparing PSEN1 mutation to genetic controls, including CRISPR engineered isogenic control lines. In addition, we will assess the impact of antibody therapies that are approved for PSEN1 mutation carriers, to determine whether they can bind A species, offset other signs of pathology, and offer insight pertinent to drug-related vascular side-effects. We will demonstrate independent replication of the model in a different institution lab to demonstrate robustness. Transparent sharing of detailed protocols and data will aid adoption of the model.
