High numerical aperture objective

The laser beam (10W, 1064 nm) was tightly focused through a series of Keplerian beam expanders and a high numerical aperture objective (100 × 1.45, oil, Nikon). A high-resolution quadrant detector was used for position detection. To measure mechanical properties of the cytoplasm, 0.5-μm diameter latex particles (Sigma, L9654) were embedded in the gel and were endocytosed by the cells as they grew into cancer organoids. The linear region of the detector and the trap stiffness were calibrated with the same bead using an active power-spectrum method and equipartition theorem^{31 (link),51 (link),52} . The endocytosed bead was dragged at a constant velocity of 0.5 μm/s by the optical trap, and the force-displacement curve of the local cytoplasm was recorded. To calculate the apparent modulus (E_A), the force and displacement were normalized by the cross-section area and the diameter of the particle respectively. The slope in the linear range of the normalized force-displacement curve was taken as the apparent modulus.

The measurement of cell cortical stiffness by optical tweezer was adapted from a well-established method in the literature^{27 (link)}. The laser beam (10 W, 1064 nm) was tightly focused through a series of Keplerian beam expanders and a high numerical aperture objective (100×/1.45, oil immersion, Nikon). A high-resolution quadrant detector (PDQ80A, Thorlabs, Newton, NJ, USA) was used for force measurement. To measure the mechanical properties of the cell cortex, polystyrene particles (500 nm in diameter, orange fluorescence) were added to the culture medium and endocytosed by the cells. The particle was then dragged by optical tweezers toward the cell membrane to deform the cell cortex with a speed of 1 μm/s. The displacement of the particle and the resistant force were recorded by the quadrant photodetectors. The stiffness of the cell cortical structure including plasma membrane and cell cortex was defined by the slope of the force-displacement curve. The data collection and post-processing were performed using MATLAB (Mathworks, Natick, MA, USA).