Raman Spectroscopy Imaging in Evaluation of Immune Cells Activation Status

Introduction: The ability to assess the activation status of immune cells is useful for medical diagnostics and currently also for monitoring the performance of immune cells-based advanced therapy medicinal products (ATMP) such as the chimeric antigen receptor (CAR)-T cells. CAR-T cells numbers and immunophenotype are routinely monitored to avoid poor expansion or exhaustion phenotype. In this study, we utilized Raman spectroscopy (RS) as a label-free, non-destructive method for assessing T cells activation and macrophages polarization (pro-inflammatory M1 vs immunosuppressive M2). RS is a well-defined, non-elastic light scattering-based spectroscopic technique that followed by chemometric analysis provides unique and comprehensive information regarding the molecular structure, metabolic signature and biochemical composition of tissues and cells at subcellular level. To assess if RS would be a valid approach for detection of CAR-T activation we first developed CD19CAR-transduced Jurkat cell line model. Although Jurkat cells, a leukemic T cell line, lack cytolytic activity, they do produce IL-2 and upregulate CD69 upon activation like CAR-T thus serve as a convenient model.

Materials and Methods: T cells and monocyte were isolated from healthy donors' blood. T cells were activated with coated CD3/CD28 beads. Jurkat cells were transduced with anti-CD19 chimeric TCR. CAR-J cells activation upon contact with CD19+ cells were confirmed by CD69 expression (flow cytometry) and calcium fluxes (fluorescence microscopy). Monocyte were differentiated to macrophages using CSF1, while THP1 monocytic cell line with PMA. Macrophages were subsequently polarized into M1 and M2 with IL-4 and IL-13 or IFN-gamma and LPS respectively. M0 macrophages or naïve T cells were used as a reference and control. Cell phenotypes were confirmed using flow cytometry, q-PCR, fluorescent confocal microscopy and proliferation assay. Cells were fixed in 0.5% glutaraldehyde and imaged at 533 and 633 nm with a WITec Alpha 300 Raman microscope. Collected data were analyzed by chemometric methods: K-means cluster analysis (KMCA) followed by Partial Least Squares-Discrimination Analysis (PLS-DA) and Principal Component Analysis (PCA).

Results: RS analysis of polarized macrophages showed the increased signal from hemoproteins (740-760 cm ^-1) in all three states. Interestingly we observed different hemoproteins to lipids (1440, 1240, 1660, 2850 cm ^-1) ratio between M0, M1 and M2. Also degree of lipid unsaturation can be considered an important parameter in the identification of the M1 phenotype of macrophages.

Chemometric analyzes revealed significant morphological (increased cell and nucleus sizes) biochemical and difference between activated T lymphocytes and naive cells. Carotenoids present in non-activated T cells disappeared after activation. Conversely, lipids in the form of distinct clusters were observed after activation. Moreover, in the spectra of cell nuclei (733, 790, 1583 cm ^-1) of activated lymphocytes, less condensed chromatin was manifested by a lower intensity of marker bands. We observed similar changes for activated CAR-Jurkat, though the morphological changes were more subtle and carotenoids were not observed in Jurkat cells.

Conclusions: The obtained results indicate the potential utility of Raman spectroscopy methods in identifying lymphocyte states activation and macrophage polarization .

Acknowledgements. The studies were performed as a part of the “ Label-free and rapid optical imaging, detection and sorting of leukemia cells” project carried out within the Team-Net programme of the Foundation for Polish Science co-financed by the European Union under the European Regional Development Fund.