Cellular discrimination using in vitro Raman micro spectroscopy: the role of the nucleolus.

International audienceRaman micro spectroscopy has attracted considerable attention over the last few years to explore its possible clinical applications as a non-invasive powerful label-free in vitro screening tool in cancer diagnosis and monitoring, subcellular analysis of biochemical processes, drug uptake, mode of action and mechanisms of interaction as well as toxicity of, for example, chemotherapeutic agents. However, in order to evaluate accurately the potential of Raman micro spectroscopy for such applications it is essential to optimise measurement and data processing protocols associated with subcellular analysis. To this end, in vitro differentiation of cell lines is a basic proof of concept for the potential of the technique, and although many studies have indicated successful differentiation based on Raman micro spectroscopy, it is important, as the measurement and processing techniques are improved, to establish the biochemical and subcellular basis of that discrimination. In this study, Raman micro spectroscopy is used to compare and differentiate normal and cancer cells from human lung origin, A549 adenocarcinoma cell line, Calu-1 epidermoid non-small-cell and BEAS-2B normal immortalized bronchial epithelium cell line. Spectra were taken from the three subcellular compartments, cytoplasm, nucleus and nucleolus and Principal Components Analysis was used to compare the spectral profiles between the cell lines and, coupled to Linear Discriminant Analysis, to explore the optimum sensitivity and specificity of discrimination. To support the analysis, Raman micro spectroscopy was coupled with Flow Cytometry, Confocal Laser Scanning Microscopy and Atomic Force Microscopy. While all subcellular regions can be employed to differentiate the normal and cancer cell lines, optimum discrimination sensitivity and specificity is achieved using the spectra from the nucleolar region alone. Notably, only the nucleolar spectral profiles differentiate the two cancer cell lines. The results point to the importance of the nucleolar regions in diagnostic applications of Raman microscopy as well as further applications in subcellular analysis of cytological processes