Chromatin fractionation related to cell type and chromosome condensation but perhaps not to transcriptional activity.

A chromatin fractionation procedure has been developed that involves the shearing of swollen chromatin and the separation of chromatin fragments by sucrose gradient centrifugation. When chromatin from rapidly growing HeLa cells was fractionated, four partially resolved peaks were obtained. Partial characterization of the chromatin fractions indicated that each contained similar lengths of DNA formed in a complex with histones in nucleosomes. The most likely difference between the fractions is the degree of intra-or interstrand association of the fibers of nucleosomes. Chromatins from chicken erythrocytes and mitotic chromosomes were analyzed by the procedure. These chromatins yielded gradient profiles that were distinctly different from each other and from interphase chromatin. These results suggested that the fractionation procedure reflected at least some of the differences in the structure of chromatin known to exist in vivo. The association of pulse-labeled RNA with a particular chromatin fraction is frequently used to support claims of successful separation of transcriptionally active chromatin from inactive chromatin. Since our data show that the slowly sedimenting fractions preferentially bind RNA whose synthesis was clearly not in progress at the start of the fractionation, this criterion is suspect. The presence of equal amounts of satellite DNA in all fractions of mouse L-cell chromatin indicated that the method did not fractionate on the basis of the in vivo transcriptional activity.