The remarkable instability of replication loops provides a general method for the isolation of origins of DNA replication.

Closed superhelical replicative intermediates of SV40 DNA were stable during incubation at elevated temperatures for prolonged times. However, when the replicative intermediates were nicked or made linear by digestion either outside or inside of the replication loop, the nascent DNA strands were spontaneously extruded at a measurable rate even at 37 degrees C. The extruded DNA was double-stranded. The rate-limiting step in the extrusion was not the rate of branch migration. The instability of replication loops suggested a method for the isolation of replication origins. The method has been tested by a reconstruction experiment in which a small amount of linear SV40 replicative intermediates was mixed with a vast excess of cellular DNA. The mixture was heated overnight at 55 degrees C, and the spontaneously extruded nascent strands were isolated and inserted into pBR322. Plasmids containing SV40 DNA of less than genomic size, symmetrically distributed about the SV40 replication origin, were isolated. This result suggests that the method is generally applicable to the isolation of replication origins.