Spatially Resolved Laser Tweezers Raman Spectral Snapshots Reveal Single-Cell Molecular Heterogeneity

Tumor heterogeneity is the major driver of cancer drug resistance, but its quantitative in situ characterization remains difficult to probe at the single-cell level. Here, laser tweezers Raman spectroscopy (LTRS) is employed to achieve micrometer-scale, spatially resolved, label-free biochemical imaging of eight representative cancer cell lines (Jurkat, Raji, NCI-H82, HCC827, PC-9, NCI-H1437, DLD-1, and CHO). Two-dimensional Raman spectral snapshots of optically trapped single-cell lines were acquired, from which pixel-resolved Raman spectra were extracted. Characteristic Raman peak ratios associated with lipids, proteins, nucleic acids, and membrane components were mapped to reveal intracellular biochemical distributions. Based on the spatial profiles of these ratios, the spatial full width at half-maximum (fwhm) was introduced as a quantitative spatial heterogeneity marker. Hematological cancer cells (e.g., Jurkat) exhibit multipeaked distributions with narrow spatial fwhm values (∼2.0–5.3 μm), reflecting organelle-level fragmented heterogeneity, whereas solid tumor cells (e.g., NCI-H82) display single-peaked profiles with broader values (∼18.6–35.9 μm), indicating more uniform and continuous molecular organization. Furthermore, a label-free classification model based on characteristic Raman ratios enables accurate tumor phenotype discrimination. This work establishes spatial fwhm-based, spatially resolved LTRS analysis as a robust framework for quantifying single-cell heterogeneity with potential applications in early cancer screening and therapeutic monitoring.