In their second paper on the structure of DNA*, Watson and Crick (pdf) described how DNA's structure suggests a pattern for replication:
"…prior to duplication the hydrogen bonds are broken, and the two chains unwind and separate. Each chain then acts as a template for the formation onto itself of a new companion chain, so that eventually we shall have two pairs of chains, where we only had one before." - Watson and Crick, 1953
This is called semiconservative replication.
Today we know that this is the pattern used by living cells, but the experimental evidence in support of semiconservative replication was not published until 1958. In the 5 years between Watson and Crick's suggestion and the definitive experiment, semiconservative replication was controversial and other patterns were considered.
Three hypothesized patterns were proposed:
- Semiconservative - The original double strand of DNA separates and each strand acts as a template for the synthesis of a complimentary strand.
- Conservative replication - the original double strand of DNA remains intact and is used as a template to create a new double stranded molecule.
- Dispersive replication - similar to conservative replication in that the original double strand is used as a template without being separated, but prior to cell division, the strands recombine such that each daughter cell gets a mix of new and old DNA. With each round of replication, the original DNA gets cut up and dispersed evenly between each copy.
Knowing what we know now about how DNA behaves, the fact that the dispersive pattern was a popular seems odd, but this pattern was favored by several well known, prominent scientists. These scientists did not like the semiconservative pattern. They thought the helical nature of the double stranded DNA molecule would make it difficult for the strands to be unwound, separated and copied in the way needed for semiconservative replication to be possible. The dispersive pattern of cutting the helix once every rotation eliminated the need for unwinding the helix. Support for the dispersive hypothesis remained strong until proof of semiconservative replication was provided by Meselson and stahl's 1958 paper
The methods Meselson and Stahl developed allowed them to distinguish existing DNA from newly synthesized DNA and to track new and old DNA over several rounds of replication.
They accomplished this by labeling cells with different stable isotopes of nitrogen. First, a culture of bacterial cell were grown for several generations in a media containing only 15N ( a stable, heavy isotope of Nitrogen). After this period* of growth, all of the DNA in the cells contained 15N. These cells were then rinsed and put into a media containing only the more common, lighter isotope of nitrogen (14N). As the cells grew and divided in this fresh media, all newly synthesized DNA would contain only the lighter nitrogen isotope, while DNA from the original cells would still contain 15N. In this illustration above, 15N labeled DNA is shown in orange and 14N labeled in green.
The 15N and 14N labeled DNA was then tracked using high speed centrifugation and a density* gradient created with cesium chloride (CsCl).
During centrifugation in a CsCl gradient, DNA accumulates in bands along the gradient based on its density. Since 15N is more dense than 14N, 15N enriched DNA accumulates lower down in the centrifuge tube than the 14N DNA. DNA containing a mixture of 15N and 14N ends up in an intermediate position between the two extremes.
By spinning DNA extracted at different times during the experiment, Meselson and Stahl were able to see how new and old DNA interacted during each round of replication.
The beauty of this experiment was that it allowed them to distinguish between the three different hypothesized replication patterns. The key result occurs at the second generation when all three proposed replication patterns give different results in the CsCl gradient.
That Meselson and Stahl's experiment showed the pattern predicted by the semiconservative hypothesis provided the definitive experimental evidence in support of the process proposed by Watson and Crick.
- Problem Sets