Automated segmentation and tracking of mitochondria in live-cell time-lapse images

Abstract

Mitochondria display complex morphology and movements, which complicates their segmentation and tracking in time-lapse images. Here, we introduce Mitometer, an algorithm for fast, unbiased, and automated segmentation and tracking of mitochondria in live-cell two-dimensional and three-dimensional time-lapse images. Mitometer requires only the pixel size and the time between frames to identify mitochondrial motion and morphology, including fusion and fission events. The segmentation algorithm isolates individual mitochondria via a shape- and size-preserving background removal process. The tracking algorithm links mitochondria via differences in morphological features and displacement, followed by a gap-closing scheme. Using Mitometer, we show that mitochondria of triple-negative breast cancer cells are faster, more directional, and more elongated than those in their receptor-positive counterparts. Furthermore, we show that mitochondrial motility and morphology in breast cancer, but not in normal breast epithelia, correlate with metabolic activity. Mitometer is an unbiased and user-friendly tool that will help resolve fundamental questions regarding mitochondrial form and function.