The morphology of mitochondria can inform about their functional state and, thus, about cell vitality. For example, fragmentation of the mitochondrial network is associated with many diseases. Recent advances in neuronal imaging have enabled the observation of mitochondria in live brains for long periods of time, enabling the study of their dynamics in animal models of diseases. To aid these studies, we developed an automatic method, based on supervised learning, for quantifying the degree of mitochondrial fragmentation in tissue images acquired via two-photon microscopy from transgenic mice, which exclusively express Enhanced cyan fluorescent protein (ECFP) under Thy1 promoter, targeted to the mitochondrial matrix in subpopulations of neurons. We tested the method on images prior to and after cardiac arrest, and found it to be sensitive to significant changes in mitochondrial morphology because of the arrest. We conclude that the method is useful in detecting morphological abnormalities in mitochondria and, likely, in other subcellular structures as well.
Keywords: Beta regression; image analysis; intravital imaging; mitochondrial fragmentation; mitochondrial morphology; two-photon microscopy.
© 2015 The Authors Journal of Microscopy © 2015 Royal Microscopical Society.