Application Scientist Molecular Devices, LLC, Austria
Abstract: Cardiovascular disease is a leading cause of death worldwide. Cardiac organoids replicate the complex architecture and function of the human heart and making them a promising tool for disease modeling and drug development However 3D model systems are often suffering from object to object variability and low throughput. This limits the use of organoids in upscaling and compound screening. Culturing in the lab can be complex, labor intensive and prone to human error. To exceed these obstacles we developed a fully automated cardiac organoid workflow performed on CellXpress.ai. We automated full workflow that starts from iPSC culture followed by differentiation, plating cells into 96 low attachment plates allowing formation of 3D cardioids, then allowing maturation of cardioids during several weeks with periodic media exchanges every two days. The process ends with an integrated end-point assay, kinetic calcium imaging done by FLIPR instrument. In parallel the cardioids were monitored on a daily basis and the data was processed with a label-free phenotypical classification analysis based on a deep learning model. Key findings of the automated cardioid workflow: The workflow using unified incubation, liquid handling, feeding and imaging. AI assisted automated decision making to initiate events like media exchange or compound treatment. Integrated software executes workflows, acquires images and simplifies the analysis of the generated data. Automated quality control includes phenotypical classification and sample tracking with the Cell Journey feature throughout the entire life span from iPSCs to Cardioids. Integrated FLIPR Penta device records and analyze calcium oscillations based on calcium dye treatment. The data here demonstrates the advantages in the automated generation, maintenance and analysis of cardiac organoids. CellXpress.ai cell culture system supports scientists in producing reliable cardioids for high throughput , personalized medicine, drug development and disease modeling by avoiding high organoid to organoid variability and human prone errors.
