Hyperspectral coherent anti-Stokes Raman scattering volumetric projection microscopy for assessing drug–cell interaction

As dynamic organelles reflecting cellular physiological states, lipid droplets not only provide essential substances for cellular life activities, but also their quantitative analysis is crucial for evaluating drug efficacy. Coherent anti-Stokes Raman scattering (CARS) microscopy, with imaging advantages including label-free, non-invasive, high sensitivity, and submicron resolution, is an ideal tool for drug efficacy assessment. However, traditional CARS microscopy excited by Gaussian beams requires prolonged focusing for 3D tomography, which is time-consuming, causes significant photodamage, and easily alters the cellular microenvironment, affecting the accuracy of quantitative analysis of lipid droplet proportion. We adopted dual Bessel beams to excite CARS signals, increasing the depth of field by 5 times and resolution by 1.17 times, enabling 3D information and volumetric Raman spectra acquisition with a single 2D scan. Detecting HeLa cells treated with gradient concentrations of doxorubicin hydrochloride via spectral phasor segmentation revealed reduced lipid droplet accumulation correlated with drug effect, supporting rapid quantitative visualization of lipid droplets and research on drug-cell interactions.