Using time-lapse confocal microscopy for analysis of centromere dynamics in human cells

Using these methods we have shown that the alpha-satellite domain of the human centromere behaves as an elastic element, stretching in response to spindle forces applied during prometaphase and metaphase (Shelby et al., 1996). These data complement previous observations of centromere stretching during mitosis (e.g., Skibbens et al., 1993) by demonstrating a specific molecular compartment within the centromere, the satellite heterochromatin domain, that supports this strain. Centromere stretching reports on the net force applied across the centromere during mitosis and the availability of a fluorescence-based assay system in human cells provides a robust assay system to complement the elegant DIC-based methods that have been perfected using marsupial and newt cell cultures (Cassimeris et al., 1990; Skibbens et al., 1993, 1995; Rieder et al., 1994). Current applications of this method are directed toward examining the relationship between centromere tension and microtubule dynamics using pharmacological approaches and the behavior of kinetochore-associated regulatory proteins, such as Mad2 and Bub1 (Li and Benezra, 1996; Chen et al., 1996; Taylor and McKeon, 1997), as a function of centromere distortion. In addition, GFP-labeled centromeres can be observed during interphase, providing a novel window into chromosome organization within the nucleus. Our observations show that centromeres distribute into the newly forming nucleus at telophase by what is apparently a uniform isometric expansion, with little evidence for directed motion of individual centromeres contributing to the formation of the G1 nucleus. During interphase, centromeres show very little movement in general, behaving as though embedded in a rigid matrix. Sustained movements of individual centromeres or groups of centromeres are occasionally observed, however, suggesting that chromosome position is subject to change during interphase (Shelby et al., 1996). These experiments complement those described by Belmont and colleagues, who have developed a method to mark specific chromosomal sites with GFP for analysis in vivo (see Chapter 13; Robinett et al., 1996; Straight et al., 1996). These new GFP-based techniques for direct observation of defined DNA sequence domains in vivo carry the logic of in situ hybridization analysis into living cells and, coupled with new methods for observing global chromatin architecture as well as functional nuclear protein domains (Huang et al., 1997; Misteli et al., 1997), promise significant progress toward understanding the dynamic organization of the genome within the living nucleus.