Authors

Discussion

Alzheimer’s disease (AD) accounts for 60-70% of dementia cases making it the most common form. Current treatments are only partially effective at alleviating AD symptoms. Novel treatments have questionable safety/efficacy. Therefore there is a need for alternative strategies e.g. drug repositioning/repurposing which can identify safe and tolerated compounds that may be effective. Age is the greatest risk factor for AD, however genetic variants are important for defining risk-level and supporting early-onset detection for individuals in middle- and old.-age. Genome-wide association studies repeatedly show variants in SORL1 (encodes the protein SORLA) increase risk for AD. Variants that confer loss of, or decreased, SORLA expression is also linked with AD. Deletion of SORL1 is associated with enlarged endosomes in neural progenitor cells and neurons specifically in keeping with the role of SORLA in the endolysosomal pathway. Therefore image-based phenotyping may identify features characteristic of SORLA depletion. Here, an automated morphological profiling assay (Cell Painting) was applied to wild-type and SORL1^-/- NPCs. This was used to determine the phenotypic response of SORL1^-/- NPCs after treatment with a preliminary drug screening library (330 compounds) to test the utility of such a study design for a future high throughput drug screen. Using multiparametric, image-based phenotypes in SORLA depleted NPCs with reference to isogenic wild-type controls, 17 FDA-approved compounds that reverse the morphological signatures in NPCs derived from 3 SORL1^-/- hiPSCs were identified. Network pharmacology analysis reveals the 17 compounds interact with a number biological pathways including DNA damage. image-based phenotyping and morphological profiling in SORL1^-/- NPCs using high-content image analysis is a viable option for drug discovery in AD.