The replication of chromosomes and individual genes is under cell-specific and developmental control. The nonrandom activation of replication origins and the asymmetry of chromatin replication at the molecular level have engendered the hypothesis that the potential for transcription of a gene is determined by the origin used to replicate that gene in the immediately preceding cell cycle. A novel method for determining replication polarity (the run-off replication assay) will be employed to test this hypothesis by comparing the direction of replication of specific genes in different states of transcriptional activity. The replication polarity of the following genes will be examined: (a) the alpha- and beta-globin genes of mouse erythroleukemia cells before and after induction by DMSO, (b) the alpha- and beta-globin genes of chick embryo fibroblasts before and after activation by Rous sarcoma virus transformation, and (c) the translocated (active) and nontranslocated (inactive) alleles of the c-myc cellular transforming gene in human Burkitt lymphoma cells. Discovery of a correlation between gene activity and replication polarity in these oncogenically transformed cell types would provide insight into both the mechanisms and cellular structures involved in transcriptional control in normal and tumor cells. Future application of the run-off replication technique will allow isolation of bona fide replication origins based on in vivo activity. The run-off replication technique involves completion of in vivo initiated nascent DNA chains in vitro, in the presence of the density labeled nucleotide bromodeoxyuridine triphosphate. The polarity of replication through a given segment of DNA is deduced after resolution of restriction enzyme generated DNA fragments by isopychic centrifugation, gel electrophoresis and blot hybridization to radiolabeled probes; those fragments most enriched in the density label being the farthest from the origin of replication. Preliminary experiments with the run-off replication technique demonstrate the faithful elongation of in vivo initiated chains in vitro, the sensitivity of detection of this method, and suggest that a correlation between replication polarity and gene activity does exist; thus the inactive avian histone H5 gene replicates from a downstream origin while the active H5 gene, and the active or inactive alpha-globin genes, replicate from upstream origins. Replication from an upstream origin therefore may be a necessary, but not sufficient, condition for transcription.
