Automated contour analysis of multi-cellular spheroids spreading through high content imaging

The morphology of 2D cell colonies has been studied to understand tumor metastasis in the past decades. However, 2D cell cultures are lacking many features of 3D tissues, and their physiological behaviors are quite different from solid tumors in vivo. In this work, we studied the multi-cellular tumor spheroid (MCTS) spreading on the substrate, which keeps parts of 3D tissue characteristics and facilitates cell tracking through 2D imaging. By using a high content imaging system (HCS), we tracked multiple spheroids in one single 96-well plate for 36 h. An automated algorithm based on Otsu's method was developed to investigate the morphological details of spheroids through the quantification of radius length and its coefficients of variation. Spheroid spreading is altered by the PIP-platin, which was a novel platinum based drug previously reported by us with an inhibitory effect on cell migration. All parameters showed dose dependent decreases when PIP-platin concentration increased, indicating the inhibition of spheroid expansion by this compound. To investigate the surface roughness of spheroids affected by the drug, we applied the Fourier parameter β and the normalized standard deviation of the radius STD_r / $⟨r⟩$, which were found inversely proportional to the concentrations of PIP-platin. Particularly at the low drug concentrations, the indices of contour roughness appeared to be more sensitive than spheroid sizes, which could be the potential morphological markers for high content screening of drugs. (paper)