No CrossRef data available.
Published online by Cambridge University Press: 31 January 2003
Abstract: It has long been debated whether the 30-nm fiber of chromatin is packed in an orderly array. The fiber may be condensed by supercoiling, producing structures of varying diameter. Alternatively, technical problems may have prevented the detection of higher-order structures. We developed a strategy to distinguish between these two possibilities. One potential obstacle to studying the order of packing was the effect of fixatives, dehydrating agents, heat, and embedding polymers on the native structure of chromatin prepared for viewing by electron microscopy. The known tendency of proteins to be degraded by osmium tetroxide and subsequently to be extracted in the conventional protocols for embedding might be particularly damaging. To avoid such denaturants and ensure the retention of proteins in chromatin, the embedding resin HACH was employed. Drosophila mimica polytene chromosomes were thin sectioned, stained with uranyl acetate, and viewed in the transmission electron microscope. Images were digitized and subjected to computerized image processing. Raw data files, containing boundary coordinates of all closed figures in the image, were edited to retain only those regions of interest (ROIs) that exhibited dimensions similar to those of 30-nm fibers in projection views. Euclidean distances between the centroids of such structures were calculated to obtain linear intercepts between recognizable 30-nm fibers. According to stereology theory, the dimensions of a lamellar structure can be determined from the volume distribution function of such intercepts. Therefore, intercept values were pooled for final data files from five processed images of chromatin. The resulting frequency histogram, showing the number of observations at different intercept values, had a sigmoidal inflection that was diagnostic of a major, new spacing at 95 nm. The 95-nm minimum was sandwiched between maxima in the 85 to 90 nm interval and throughout the range 105 to 120 nm. The results suggest that established stereological theory will be a useful tool for investigating the intractable problem of higher-order chromatin structure.