高度多重成像实现肿瘤类器官模型的空间表征

The emergence of 3D organoid models bridged the gap between 2D cell culture models and complex animal models. Its ability to accurately recapitulate the structure and function of organs and tumors made it a suitable system to perform mechanistic and perturbation studies with more complexities than a 2D cell culture but without the inconsistencies of animal models mimicking human diseases. One of the advantages of organoid models is it accurately represents a tissues structural environment. However, methods to image and spatially characterize organoid models have typically been limited to H&E staining which only gives a qualitative aspect of the structure. While single-plex immunohistochemical analysis is also possible, these tissues are limited and can be exhausted before full evaluation on multiple slides. Having the ability to image and characterize these structures in an efficient way not only validates the organoid's effectiveness to recapitulates the environment of the original tissue but also highlights spatial changes in these models throughout an experiment using only a single tissue slide. For this study, we produced patient-derived lymphoma and glioblastoma (GBM) cell line derived organoids. Organoids were fixed in hydroxyethyl agarose gel and embedded in paraffin. To spatially characterize the organoids produced, we utilized an automated sequential immunofluorescence imaging platform for its ability to detect up to 80 different proteins in a single protocol run. We selected 50 markers to develop the multiplex panel for organoid profiling. The panel consists of markers targeting extracellular matrix, cell phenotypes (GBM and lymphoma), and metabolic state. We believe that these set of markers can characterize and confirm the recapitulation of the two different organoids. In the patient derived lymphoma organoid, we observed cell type frequencies of T-cell subsets and B-cells which recapitulated flow cytometric analysis of the same organoid, confirming its ability to find relevant cell populations. This was additionally true for the GBM organoid, which contained a mixture of different neural cell states, similar to low-complexity single-plex assays performed on the same organoid. In addition, proliferative T-cells in the lymphoma organoid system also had higher levels of glycolytic and mitochondrial enzyme expression compared to non-proliferative subsets. This preliminary proof of principle demonstrates the utility of highly multiplexed imaging for the evaluation of organoid models with broad implications for pre-clinical and translational studies.