The most beautiful experiment in biology – The Meselson-Stahl experiment

Here is the example of scientific beauty I promised last week. Most people will agree right away that the Mona Lisa is a great painting. You don’t need to study it in depth to realize that, though it might be worth a second glance to appreciate more of its beauty. As is true with most art. The same however is true for science. Let’s look at what some call biology’s most beautiful experiment:

The golden era of DNA research

It was published in 1958, in the golden era of DNA research. Oswald T. Avery had just shown some years ago that DNA is the hereditary material, and a mere 5 years ago (just yesterday in scientific terms) Jim Watson and Francis Crick published their famous paper on the structure of DNA. Along with their model that shows the double helical DNA molecule consisting of two antiparallel DNA strands (that means that the two strands lie “head to toe” to each other) they also developed an idea how the DNA, our genetic material, might be copied. They concluded their paper with the famous sentence that “It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.”, one of the most famous sentences in science for all the understatement it contains.

The problem

However, their suggested mechanism of DNA replication was not the only one out there. Various influential scientists suggested 3 different theories; future Nobel Prize winners were championing one or the other. The whole world was discussing whether DNA would get replicated in a conservative, a semi-conservative or a dispersive fashion (ok, rather a few expert scientists and not the whole wide world, but still…). Conservative replication means that one double stranded DNA molecule would stay together all the time and would produce a copy of 2 new strands. Semi-conservative replication means that the strands would separate during replication and both daughter molecules would contain one of the original strands and a new complementary strand. Finally, dispersive replication means that the original strands would get broken up into smaller pieces that get individually copied and then the two daughter molecules would each contain a mixture of old and newly synthesized pieces of DNA. No one could come up with a working experiment to prove one or the other up to 1958.

Image modified from the original image by Adenosine,

The idea

The stroke of genius that Matthew Meselson and Franklin Stahl came up with to solve this problem is beautiful in its simplicity. They were aware that they had to find a way to distinguish the old strands from the newly made ones. DNA can be readily detected by UV light but you cannot distinguish between freshly made and old DNA. When they realized that you could float different things in solutions of a different density, the idea for their experiment was born. It is based on the same principle that allows us to float in the water of the Dead Sea between Jordan and Israel but not in regular seawater.

The water of the Dead Sea contains much more salt than other seas. Thus water from the Dead Sea has a higher density than our average water and we can float on it. Meselson and Stahl adopted the same principle for their experiment and developed a technique called density gradient centrifugation. Basically, they put salty water in a centrifuge and spun it at very high speeds. The high-speed centrifugation forms a gradient in the salty solution with a higher density at the bottom of the tube, as the heavy atoms will gradually move there. Any molecule placed in the solution will eventually end up at the region that has the same density as the molecule itself.

What was left before starting the experiment was to find a good salt with the right density for DNA (they found Cesium Chloride to work well) and a good way to make one sample of DNA heavier than the other one. After some trial and error they succeeded at labeling DNA with a heavier isotope of nitrogen,15N . All that is important here is that this makes the DNA molecule significantly heavier than regular DNA containing the usual 14N isotope, as DNA contains a lot of nitrogen. They also had access to a shiny new ultracentrifuge in which you they could separate the heavy from the light DNA and take a picture of the samples while spinning them in the centrifuge.


They reasoned the following: If we grow bacteria in a solution containing only the heavy nitrogen and none of the regular one, after a while, all the DNA in the bacteria will contain almost exclusively heavy nitrogen. If we then take them out of this growth medium and into a regular one containing the regular nitrogen, all newly made DNA will contain only the light nitrogen. Let’s separate the DNA molecules that we obtain from the bacteria after several generations of bacterial growth on our density gradient (a generation is the time the bacteria need to double). We should observe bands of DNA on the pictures at the density of the DNA molecules. Lighter DNA should float higher than heavier DNA, this we can quickly confirm with control heavy and light DNA. Depending on the patterns of heavy and light DNA we will observe in the experiment, we should be able to distinguish between the different models of DNA replication.

Can you tell how?

The result

The result was clean as a whistle. At the start of the experiment before shifting to the regular medium, they saw only one band of DNA on the pictures of their density gradient. It consisted exclusively of heavy DNA as no growth in the regular medium had occurred yet. After one bacterial generation they still only observed one band although at a lower density in the gradient (i.e. higher up!), about half way between control heavy and control light DNA! This means that DNA replication cannot be conservative, as conservative replication would lead to two distinct bands, one consisting exclusively of heavy DNA and one exclusively of light DNA. But that still leaves options 2 and 3. Yet another round of bacterial growth gave the final answer: They now saw two distinct bands, one at the density of the control light DNA and one at the intermediate density from last round. This excludes dispersive replication as a mechanism as dispersive replication would have resulted in yet another single band at a density between the control light DNA and the control heavy DNA.

Meselson-Stahl experiment

Image by LadyofHats

DNA replication is semi-conservative! Jim Watson and Francis Crick who had proposed the semi-conservative model had been right (again).

This discovery was so important for biology because it focused the thinking of many bright minds on the right mechanism of DNA replication rather than having their attention divided. It provided the basis for many more in-depth discoveries on the replication machinery and the molecular players involved.

All it took was one very simple and elegant experiment to solve the mystery of the mechanism of DNA replication that had been elusive before. I don’t know wheter it is the most beautiful biology experiment of all time, but it sure is an example of true scientific beauty.



the original 1958 paper in PNAS

This entry was posted in science and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s