The interaction of the linker histones Hl and H5 from chicken erythrocyte chromatin with pBR322 was studied as a function of the number of superhelical turns in circular plasmid molecules. Supercoiled plasmid DNA was relaxed with topoisomerase I so that a population with a narrow distribution of topoisomers, containing from zero to five superhelical turns, was obtained. None of the topoisomers contained alternative non-B-DNA structures. Histone-DNA complexes formed at either 25 or 100 mM NaCI fmal concentration and at histone-DNA molar ratios ranging from 10 to 150 were analyzed by agarose gel electrophoresis. The patterns of disappearance of individual topoisomer bands from the gel were interpreted as an indication of preference of the linker histones for crossovers of double-helical DNA. This preference was observed at both salt concentrations, being more pronounced under conditions of low ionic strength. Isolated H5 globular domain also caused selective disappearance of topoisomers from the gel, but it did so only at very high peptide-DNA molar ratios. The observed preference of the linker histones for crossovers of double-helical DNA is viewed as a part of the mechanism involved in the sealing of the two turns of DNA around the histone octamer.The lysine-rich histones (Hi and its variants) interact with the linker DNA between nucleosomes, sealing two turns of DNA around the nucleosome core (1). They are also involved in forming higher-order structures of the chromatin fiber (for reviews, see refs. 2 and 3). Recent evidence shows that the lysine-rich histones are involved in the regulation of gene transcription (reviewed in refs. 4-7).One of the main features of the interaction of Hi and DNA is the preference of this histone for superhelical DNA. However, the original work from Singer's laboratory (8-10), which indicated that Hi had a higher affinity for superhelical than for relaxed circular or linear DNA, has been questioned on a number of occasions (see ref. 6). While direct competition experiments confirmed the claimed preference for superhelical DNA molecules in the case of Hi (11,12) and H10 (12), the issue of whether histone H5 [the Hi variant specific to nucleated erythrocytes (13, 14)] possesses such a property is still a matter of controversy (11,15).Previous studies of this phenomenon are subject to a general criticism. In most cases plasmid or viral DNA preparations were used that were poorly characterized with respect to their topological state. Either total populations of closed circular DNA were used directly as isolated from bacterial or eukaryotic cells or high levels of supercoiling were induced by incubating nicked circular molecules with ethidium bromide and then ligating or treating with topoisomerase in the presence of the intercalator. In such preparations the superhelical density is not precisely known, and it isThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U...