To better understand the impact of changes in nuclear architecture on nuclear functions, it is essential to quantitatively elucidate the three-dimensional organization of nuclear components using image processing tools. We have developed a novel image segmentation method, which involves a contrast enhancement and a subsequent thresholding step. In addition, we have developed a new segmentation method of the nuclear volume using the fluorescent background signal of a probe. After segmentation of the nucleus, a first-order normalization is performed on the signal positions of the component of interest to correct for the shape of the nucleus. This method allowed us to compare various signal positions within a single nucleus, and also on pooled data obtained from multiple nuclei, which may vary in size and shape. The algorithms have been tested by analyzing the spatial localization of nuclear bodies in relation to the nuclear center. Next, we used this new tool to study the change in the spatial distribution of nuclear components in cells before and after caspase-8 activation, which leads to cell death. Compared to the morphological TopHat method, this method gives similar but significantly faster results. A clear shift in the radial distribution of centromeres has been found, while the radial distribution of telomeres was changed much less. In addition, we have used this new tool to follow changes in the spatial distribution of two nuclear components in the same nucleus during activation of apoptosis. We show that after caspase-8 activation, when centromeres shift to a peripheral localization, the spatial distribution of PML-NBs does not change while that of centromeres did. We propose that the use of this new image segmentation method will contribute to a better understanding of the 3D spatial organization of the cell nucleus.
- Confocal Microscopy
- Image Processing