The replication of deoxyribonucleic acid (DNA) is a foundational biological process that ensures genetic information is accurately passed from a parent cell to its two daughter cells before division. Following the discovery of the double-helix structure, scientists immediately recognized that the molecule’s unique architecture suggested a mechanism for self-copying. While it was clear that DNA had to duplicate, the exact physical manner in which the two strands separated and served as templates for new strands was a subject of intense theoretical debate. Early researchers proposed multiple conceptual models to explain the mechanism of this replication, each with different predictions for how the original parental DNA material would be distributed in the resulting daughter molecules.
Defining the Conservative Model
The Conservative Model was a hypothetical proposal for how the parental DNA double helix might be copied to produce two new daughter helices. This model suggested that the original, intact parental DNA molecule would remain completely preserved after the replication event. The parental molecule would act as a template for the synthesis of an entirely new DNA double helix. The outcome of one round of replication would be two complete double-stranded DNA molecules: one molecule of the original “old” parental DNA, fully conserved and unchanged, and one molecule composed entirely of “newly synthesized” DNA strands.
Alternative Hypotheses for DNA Replication
Alongside the conservative model, two other primary theoretical mechanisms were considered for DNA replication: the semiconservative and the dispersive models. The Semiconservative Model, proposed by James Watson and Francis Crick, suggested that the two strands of the parental double helix would unwind and separate. Each original strand would then serve as a template for the creation of a new, complementary strand, resulting in two new DNA molecules, each containing one old and one new strand. The Dispersive Model proposed that the parental DNA molecule would be broken down into fragments during replication. The resulting daughter DNA molecules would then be constructed as a patchwork of short segments of both old and newly synthesized DNA, meaning every strand would be a hybrid mixture of parental and new material.
The Experiment That Ruled Out the Conservative Model
The debate between these three models was settled in 1958 by Matthew Meselson and Franklin Stahl. Their methodology relied on using nitrogen isotopes, a component of DNA bases, to physically distinguish between old and new DNA strands based on their density. They began by growing Escherichia coli bacteria in a medium containing a “heavy” isotope of nitrogen, \(^{15}text{N}\), until all of the bacteria’s DNA was labeled. The researchers then transferred the \(^{15}text{N}\)-labeled bacteria to a new medium containing the common, “light” isotope, \(^{14}text{N}\), allowing the bacteria to divide only once.
The DNA was extracted and analyzed using density gradient centrifugation. DNA containing the heavy \(^{15}text{N}\) would settle lower in the centrifuge tube, while DNA containing the lighter \(^{14}text{N}\) would remain higher. If the Conservative Model had been correct, the result after one generation would have been two distinct bands of DNA: the heavy, original parental DNA (\(^{15}text{N}\)) and the entirely new, light DNA (\(^{14}text{N}\)).
However, the actual result was a single band of DNA at an intermediate density, exactly halfway between where the heavy and light DNA would settle. This single intermediate band demonstrated that every new DNA molecule was a hybrid containing both old and new material, immediately eliminating the conservative model’s prediction of two entirely separate molecules. The hybrid result was consistent with both the semiconservative and dispersive models, requiring a second round of replication. After a second generation in the \(^{14}text{N}\) medium, the DNA produced two bands: one intermediate band and one light band, a result that perfectly matched the prediction of the semiconservative model.