Experimental Evidence of DNA Replication: Taylor's Experiment

The semiconservative model of DNA replication, though logically elegant, still needed direct experimental proof, which J. Herbert Taylor, Philip Woods and Walter Hughes supplied by working with the root cells of the broad bean, Vicia faba. They grew actively dividing root tips in a growth medium containing 3H-thymidine, a radioactively labelled precursor that gets incorporated directly into newly synthesised DNA, so that after one round of replication in this medium, the DNA of every chromosome carried the radioactive label. They then transferred the labelled root tips into an unlabelled medium containing colchicine, a chemical that arrests dividing cells at metaphase, allowing the researchers to examine each chromosome's chromatids individually at a defined stage and visualise the distribution of radioactivity by autoradiography, in which the labelled DNA exposes a photographic film to leave a pattern of scattered black silver-grain dots directly over the radioactive chromosome regions. In the first generation of chromosomes examined, the radioactive label appeared spread across both sister chromatids of every chromosome, exactly as expected if the original DNA duplex (now fully labelled on both of its two strands) had each strand serve as a template for one new, unlabelled complementary strand. In the second generation, however, only one of the two sister chromatids in each chromosome carried the radioactive label, while the other chromatid was unlabelled -- a pattern that only makes sense under the semiconservative model, in which each new duplex conserves exactly one of its two constituent stran …