(A-9) Spatially resolved 3D gene neighborhood networks for prediction of immunomodulatory potential of mesenchymal stem cells

Introduction:: Mesenchymal stem cells (MSCs) are widely studied as therapeutical cells and have shown immunomodulatory potential. MSCs are commonly isolated from bone marrow, umbilical cord, and adipose tissue. However, donor health status and tissue source differences introduce heterogeneity in MSCs, causing uncertainty in MSC potency. Many coculture-based potency predictions have been developed to examine the immunomodulatory potential of MSC populations¹. Further analysis of coculture systems is needed to better predict MSC potency. Therefore, we utilized the immune cell population in human peripheral blood mononuclear cells (PBMCs) to develop MSC-PBMC coculture system using cultrex gel (Figure 1A). We further performed sequential fluorescent in situ hybridization (seqFISH) to generate multiplexed spatial transcriptomic profiles (Figure 1C, 1D). Further, we studied gene proximity relationships using the spatially resolved gene neighborhood network (SpaGNN)² pipeline.


Materials and Methods:: PBMCs were first isolated from fresh whole blood using the Ficoll density gradient method and cryopreserved. Human bone marrow-derived MSCs were first cultured on poly-L-lysine-coated glass coverslips for 24 hours until cell adherence. The MSCs were then primed with human IFN-g diluted to 50ng/mL of culture medium for 72 hours. PBMCs were culture-rescued for 24 hours and suspended in serum-free RPMI 1640 medium. The PBMC suspension was mixed with cultrex gel at 4:1 ratio. The gel mix was then added on top of the coverslips and incubated at 37°C for one hour. Then, Ionomycin and PMA activation mix diluted in RPMI 1640 medium was added to the sample to incubate for 24 hours before fixation. SeqFISH was performed to obtain multiplexed images of RNA transcripts³ followed by immunofluorescent imaging to identify cell types².


Results, Conclusions, and Discussions:: Each image consists of a stack of images to accommodate the thickness of the sample. The positions of each RNA molecule were detected in 3D. All detected mRNA molecules were combined and clustered based on their 3D locations into regions with high local RNA density called patches (Figure 1D). Local RNA neighborhoods are detected within each subcellular region by finding the ten nearest neighbors of each RNA dot, and a nearest-neighbor network was formed by connecting RNA molecules detected as nearest neighbors (Figure 1E, 1F). We then performed a permutation analysis to compare the number of neighboring events between two genes to the expected number of neighboring events if the spatial distribution of transcripts were random. The gene proximity relationships are visualized in network format (Figure 1G, 1H). We will continue to examine gene neighborhood networks around suppressed and non-suppressed t-cells to eventually perform prediction of the t-cell suppression potential of MSCs.

In conclusion, multiplexed spatial transcriptomics can be used to examine the MSC-PBMC coculture system. SpaGNN pipeline can be used to find gene proximity relationships in 3D.


Acknowledgements (Optional): :

References (Optional): : 1. Schneider RS, Vela AC, Williams EK, Martin KE, Lam WA, García AJ. High‐Throughput On‐Chip Human Mesenchymal Stromal Cell Potency Prediction. Adv Healthcare Materials. 2022;11(2):2101995. doi:10.1002/adhm.202101995

2. Fang Z, Ford AJ, Hu T, Zhang N, Mantalaris A, Coskun AF. Subcellular Spatially Resolved Gene Neighborhood Networks in Single Cells. in press at Cell Reports Methods. doi:10.1101/2022.08.03.502409

3. Coskun AF, Cai L. Dense transcript profiling in single cells by image correlation decoding. Nat Methods. 2016;13(8):657-660. doi:10.1038/nmeth.3895