Cells of Escherichia coli containing the plasmid F were 7-irradiated with various doses to introduce determined numbers of single-strand breaks in the F DNA. The cells were then incubated to permit repair of the breaks while DNA gyrase was inhibited with coumermycin to limit restoration of any relaxed supercoils. Repaired, covalently continuous F DNA was isolated and its superhelical density was measured by two different methods. Both indicated that a major part (50-60%) of the negative superhelical turn's were maintained in the repaired molecules, suggesting that the supercoils are restrained in vivo.In eukaryotic chromatin and chromosomes, the first order of DNA packaging seems to be attained via the nucleosome structure (for reviews, see refs. 1 and 2). The double-helical DNA in this particle is coiled through its interaction with the core histones into about 13/4 toroidal superhelical turns per core nucleosome (3, 4). Prokaryotic DNA has superhelical densities similar to those found in eukaryotes (5, 6), but thereis no clear evidence for restraint of the supercoils in structures analogous to the nucleosome. Histone-like proteins are found in prokaryotes (7,8), and their mode of interaction with DNA in prokaryotic chromosomes may have some similarity with nucleosomic histones (8, 9). Also, beaded structures having the appearance of nucleosomes have been seen in an electron microscope study of DNA from partially lysed bacteria (10).However, pure DNA can spontaneously form beaded structures in certain solvents (11), and electron microscopic evidence alone cannot be considered sufficient to establish the existence of the prokaryotic equivalent of a nucleosome.DNA gyrase seems to play a vital role in forming DNA supercoils in prokaryotic chromosomes. This enzyme can catalyze an ATP-driven reaction that has the topological effect of decreasing the linking number between the two continuous strands of a DNA double helix, thereby introducing negative superhelical turns into the DNA (12). When the DNA gyrase is inhibited by coumermycin (or oxolinic acid) in intact Escherichia coli cells, X DNA molecules, which are converted in vivo to the covalently circular form, attain only a small fraction of their normal superhelical density (13,14). This result indicates that the DNA gyrase activity is necessary for the formation (or the maintenance or both) of at least part of the DNA supercoils in prokaryotic cells. Thus, it seems possible that, in viwo, supercoils in prokaryotic DNA may not be stabilized by a nucleosome-like structure and that the Major reason for both the formation and maintenance (15) of DNA supercoils is the activity of DNA gyrase. Other alternatives include the possibility that DNA supercoils catalyzed by the DNA gyrase are subsequently restrained in a nucleosome-like structure by interactions involving molecules other than DNA gyrase (9). These two alternatives differ in that in the former case the double helix (17). After irradiation, the 32p-labeled cells kept at 0C were divided into several p...