Intestinal organoids for automated screening assays. High content imaging and analysis of organoid morphology.

3D cell models representing various tissues were successfully used for studying complex biological effects, tissue architecture, and functionality. However complexity of 3D models remains a hurdle for the wider adoption in research and drug screening.

Here we describe a workflow for automation of organoid culture. The automated method utilizes an integrated work-cell, consisting of several instruments providing automated cell culture, monitoring, and a high-content imaging. The integrated system included IXM-C HT.ai confocal imaging system, automated CO2 incubator, automated liquid handler (Biomek i7), as well as Collaborative Robot. We developed methods for automation of the seeding, media exchange, as well as monitoring development of intestinal organoids. In addition, method allows automation of compound testing and evaluation of phenotypic changes.

3D intestinal organoids were developed from primary mouse intestinal cells cultured in matrigel. Cells were cultured in a media that promote the formation of 3D structures recapitulating the morphological and functional characteristics of intestine. Organoids self-organized and developed into complex structures resembling intestinal crypt formation. Developing organoids comprised objects with protrusions, cavities and vesical structures. Using automated liquid handling system allowed automated seeding cell in matrigel droplets into 24 well plates, followed by automated media addition and media exchanges. Organoids were monitored using imaging in transmitted light. Then machine learning-based image analysis allowed detection of organoids and characterization of their size, diameter, and density. For endpoint measurements organoids were then stained with fluorescently labeled antibodies or viability dyes and imaged using automated confocal imaging system. Advanced image analysis allowed by 3D reconstitution and complex phenotypic evaluation of organoid structures, including characterization of organoid size and complexity, cell morphology and viability, as well as determining presence and expression levels for differentiation markers. We characterized multiple quantitative descriptors that could be used for studying disease phenotypes and compound effects. 3D image analysis provided quantitation of the organoids number, size distribution, complexity, cell content, viability, volumes, as well as quantitation of cell proliferation and expression of specific markers. We demonstrated concentration-dependent effects of several compounds that have been known to cause toxicity (doxorubicin, cisplatin, Mitomycin C, taxol). Intestinal organoids were also evaluated for responses to stimulation with TNFa or other inflammatory cytokines.

Described methods demonstrate the tools for increase of throughput and automation in organoid assays and compound screening, and also propose analysis approaches and descriptors that allow to gain more information about the complex systems, disease phenotypes and compound effects.