摘要：

Pneumococcal DNA was labeled with high specific activities of phosphorus-32 or tritiated pyrimidines and stored at 30 °C in a casein hydrolysate medium. The kinetics of inactivation of genetic markers on the labeled DNA was examined using a transformation assay. In addition, the kinetics of DNA strand scission were examined as a function of isotope disintegration. Strand scission was determined by sedimenting labeled DNA in neutral and alkaline sucrose density-gradients containing sedimentation reference markers. Sedimentation coefficients were converted to molecular weight using Studier's (1965) relationship. The efficiency of strand scission per disintegration was then calculated. Disintegration of 32P causes single-strand breaks with unit efficiency and double-strand breaks with an efficiency of 5%. Tritium disintegration produces single-strand breaks with an efficiency of 30% and double-strand breaks with an efficiency less than 1%. The biological activity of pneumococcal transforming DNA labeled either with 32P or tritium declines as a simple exponential function of the number of disintegrations of the incorporated isotope. Heterozygous DNA hybrids containing a radioisotope in only one strand were prepared by annealing radioactive DNA with DNA labeled with the heavy isotopes 2H and 15N. The hybrid molecules were fractionated on cesium chloride density-gradients and the inactivation kinetics of genetic markers in the hybrid DNA were examined. The rates of inactivation of markers on the labeled strand were nearly equal to the rates observed with homozygous DNA labeled in both strands. From this result we conclude that single strands of transforming DNA constitute the predominant "sensitive element" inactivated by disintegration of the radioisotopes. Genetic markers on the unlabeled strand of a tritium containing hybrid are four to five times less sensitive than are markers on the labeled strand. Homozygous DNA duplexes containing 32P in only one strand appear to have multicomponent inactivation kinetics. These observations lend further support to single-strand integration models and lead us to postulate that a single-strand break constitutes an inactivating event.

展开