Structure of simian virus 40 chromosomes in nuclei from infected monkey cells

Previous analysis of the structure of isolated simian virus 40 chromosomes in this and other laboratories has suggested that they consist of nucleosomes separated by stretches of non-nucleosomal DNA highly variable in length. However, the appearance of internucleosomal DNA may have been an artifact resulting from isolation procedures and methods of analysis. To avoid such potential artifacts, we have examined the arrangement of nucleosomes in SV40^† † Abbreviations used: SV40, simian virus 40; SV40(I) DNA, closed circular, superhelical DNA; SV40(II) DNA, relaxed circular DNA containing at least 1 interruption in the phosphodiester backbone of one or both strands; SV40(III) DNA, linear, full-length DNA. chromosomes by micrococcal nuclease digestion of both viral and cellular chromatin together in nuclei isolated from SV40-infected CV-1 cells. This allowed a direct comparison between the structure of viral chromosomes and those of the host. Following electrophoresis of purified DNA digestion products, viral DNA was identified using the blotting-hybridization technique of Southern (1975), while cellular DNA was identified either by the same technique or by simply staining with ethidium bromide. While as many as 18 cellular DNA bands were detected in agarose gels, no more than six viral DNA bands were observed, and the viral DNA bands appeared broader and less well-resolved than the corresponding DNA bands from CV-1 chromatin. The average repeat distance between DNA oligomers from SV40 nucleosomes, 187 ± 11 base-pairs, was essentially the same as that observed for its host, 182 ± 6 base-pairs. These observations were independent of the extent of digestion and of the time after infection when nuclei were isolated. Since micrococcal nuclease digestion of SV40 chromosomes, both isolated and remaining in the nuclei, gave essentially the same results, the isolated chromosomes were not altered during their preparation. The results of these experiments, together with other data, have been used to construct a model for the arrangement of an average of 22 nucleosomes per genome in SV40 chromosomes that includes regions of non-nucleosomal DNA between some or all of the nucleosomes.