Analysis of Organoid and Immune Cell Co-Cultures By Machine Learning-Empowered Image Cytometry
3D organoids offer an incredibly useful tool to enhance our understanding of the intra- and inter-cellular mechanisms within tissues for medical research. However, their great potential within the field of medicine has been hindered by the major technical challenges associated with detecting and quantifying organoids.
The recently launched StrataQuest Organoid App effectively tackles these challenges by autonomously identifying organoids, enabling to monitor their growth under various conditions, and quantifying their count, dimensions, and density within co-cultures involving immune cells. This advancement facilitates in-depth investigations into intricate matters concerning the immune system, translational medicine, oncology, precision medicine, and beyond.
From 2D to 3D Cell Line Cultures
Due to their relative affordability and ease of use, the majority of human cell research has traditionally centered around 2D cell line cultures for many decades. Although they have proved successful in numerous areas of medical research, such as identifying new druggable targets, their initial formation is very inefficient, requiring extensive genetic and phenotypic alterations for them to exist in culture conditions.1,2 On top of this, almost all the differentiated cell types and tissue microenvironments present in the original tissue are absent in 2D cell line cultures, meaning these models cannot truly mimic complex in vivo processes.3 Therefore, their use in tissue studies and personalized medicine is limited.
Today, with human organoids – 3D cell line cultures derived from stem cells that contain organ-specific cell types that self-organize via cell sorting and spatially restricted lineage4 – scientists can re-create the architecture and physiology of human organs in extraordinary detail. The ability of 3D organoid cell lines to autonomously organize and accurately resemble the cellular composition and structure of human tissue makes them a highly attractive technique for medical research. They are extremely useful for analyzing therapeutical properties relating to immune disorders, organ development, oncology, host-pathogen interactions, wound healing, and drug development, and look to pave the way for precision therapy.5
Co-Cultured Organoids
Co-culturing organoids with the cells from their surrounding tissue enhances the depth of analysis for the physiological processes that occur between organoids and neighboring cell subsets, such as immune cells, stromal cells, and parenchymal cells.6 This approach allows researchers to gain insights into how adjacent cells influence organoid development, employing both time-lapse and end-point analysis techniques through the use of brightfield imaging.
However, quantifying organoid size, number and shape is still incredibly challenging because:
- Organoids, embedded in extracellular matrix gels, grow at different focal depths, making it challenging to obtain clear focus with fixed microscopy.
- Organoids vary in size and shape due to their different stages of differentiation, leading to heterogeneity.
- Dense clusters of proliferating immune cells can interfere with imaging signatures, resulting in false positives.5
Imaging-processing algorithms have mostly been developed from the analysis of organoids without additional cell subsets and until now, no high-throughput image analysis pipeline could accurately identify and quantify co-cultured organoids. 5
Examining Co-Cultured Organoids with the Organoid App
A team of scientists, therefore, developed the Organoid App to tackle these problems. The App provides reliable and efficient high-throughput validation and quantification for organoids co-cultured alongside immune cells. This breakthrough technique will allow scientists to explore highly complex questions regarding the influence of immune cell subsets and other compounds on organoid development.
The StrataQuest-supported App was created to enhance the automated identification and quantification of organoid structures in lymphocyte co-cultures. The machine learning algorithms within the Organoid App allowed for the identification and classification of 3D organoids with co-cultures of dense immune cells with extreme accuracy, mirroring that of the human eye.5
Classifier training and quality control by the StrataQuest-supported Organoid App. Source: Stüve, P., Nerb, B., Harrer, S., Wuttke, M., Feuerer, M., Junger, H., Eggenhofer, E., Lungu, B., Laslau, S. and Ritter, U. 2023. Analysis of organoid and immune cell co-cultures by machine learning-empowered image cytometry. Frontiers in Medicine. 10, p.1274482.
The StrataQuest software is capable of handling various image configurations and formats, allowing for a comprehensive analysis of co-cultured organoid development from time-lapse to end-point investigations. The end-point analysis is conducted seamlessly through automation on the TissueFAXSiPLUS microscopy platform. The method combines automated imaging with advanced image processing, including greyscale conversion, contrast enhancement, membrane detection and structure separation. 5
In particular, it includes a classifier engine used for machine learning on the samples available. A grayscale image was used as the input image, and then a user provides data on classes of objects – for example by drawing the organoid contour. This way, machine learning algorithm uses this information for more precise output.5
StrataQuest is a comprehensive image analysis software suitable for a large variety of applications, both for brightfield and fluorescence images. StrataQuest is also compatible with imaging formats of different popular imaging platforms and allows results to be extracted in different file formats including, .pdf, .xls and .xlsx, for graphical display and statistical analysis of the results.
Furthermore, the scientists compared the Organoid App with Incucyte® software and OrganoSeg, which were not capable of performing the same task as efficiently. Incucyte could not detect the differences in organoid growth between untreated and Epidermal Growth Factor-treated co-cultures, while OrganoSeg had very limited organoid detection. Thus showing the uniqueness of the effectiveness and quality of the Organoid App’s performance for such analysis.5
Conclusion
The StrataQuest-supported Organoid App provides an invaluable tool to identify and quantify organoids in cell subset cultures, which previously was not possible. This newly developed App, therefore, has the potential to completely transform medical research, especially translational medicine.
StrataQuest offers a range of customized Apps for easier navigation and analysis workflow, which are created during close collaboration with our clients. Get in contact today to see how our customized apps, including the Organoid App, can assist your research.
References
- Schutgens, F. and Clevers, H. 2020. Human organoids: tools for understanding biology and treating diseases. Annual Review of Pathology: Mechanisms of Disease. 15, pp.211-234.
- Kim, J., Koo, B.K. and Knoblich, J.A., 2020. Human organoids: model systems for human biology and medicine. Nature Reviews Molecular Cell Biology. 21(10), pp.571-584.
- Abdul, L., Xu, J., Sotra, A., Chaudary, A., Gao, J., Rajasekar, S., Anvari, N., Mahyar, H. and Zhang, B. 2022. D-CryptO: deep learning-based analysis of colon organoid morphology from brightfield images. Lab on a Chip. 22(21), pp.4118-4128.
- Lancaster, M.A. and Knoblich, J.A. 2014. Organogenesis in a dish: modeling development and disease using organoid technologies. Science. 345(6194), p.1247125.
- Stüve, P., Nerb, B., Harrer, S., Wuttke, M., Feuerer, M., Junger, H., Eggenhofer, E., Lungu, B., Laslau, S. and Ritter, U. 2023. Analysis of organoid and immune cell co-cultures by machine learning-empowered image cytometry. Frontiers in Medicine. 10, p.1274482.
- Huch, M., Gehart, H., Van Boxtel, R., Hamer, K., Blokzijl, F., Verstegen, M.M., Ellis, E., Van Wenum, M., Fuchs, S.A., de Ligt, J. and van de Wetering, M. 2015. Long-term culture of genome-stable bipotent stem cells from adult human liver. Cell. 160(1), pp.299-312.
