Measuring Superhelical Turns

Measuring Superhelical Turns

Superhelical turns in DNA may introduce distortions or torsion in the molecules that assist or hinder processes we would like to study such as recombination or the initiation of transcription. Supercoiling must be easily measurable in order to be productively studied. One way to measure superhelical turns might be to observe the DNA in an electron microscope and see it twisted upon itself. Quantitation of the superhelical 

turns in DNA with more than a few turns is difficult however. More convenient measurement methods exist. Consider the DNA molecule just described with 100 negative superhelical turns. Because it is so twisted upon itself, the molecule is rather compact and sediments in the ultracentrifuge at a high rate. If the sedimentation is performed in the presence of a low concentration of ethidium bromide, a few molecules will intercalate into the DNA. This will reduce the number of negative superhelical turns, thereby opening up the DNA, which will sediment more slowly than it would in the absence of ethidium bromide.

Figure 2.11 Sedimentation rateof a covalently closed circular DNA molecule as a function of the ethidium bromide concentration in the centrifugation solution.

Consider a series of sedimentation measurements made in the pres-ence of increasing concentrations of ethidium bromide. At higher and higher concentrations of ethidium bromide, more and more will inter-calate into the DNA and unwind the DNA more and more. Consequently the DNA will become less and less compact and sediment more and more slowly (Fig. 2.11). Finally a concentration of ethidium bromide will be reached where the molecule is completely free of superhelical turns. At this concentration, the DNA will sediment most slowly. If the centrifugation is done in the presence of still higher concentrations of ethidium bromide, the molecule will be found to sediment more rapidly as the DNA acquires positive superhelical turns and becomes more compact again. The concentration of ethidium bromide required to generate the slowest sedimentation rate can then be related to the number of superhelical turns originally in the DNA via the affinity of ethidium bromide for DNA and the untwisting produced per interca-lated ethidium bromide molecule.

Even more convenient than centrifugation for quantitation of super-helical turns has been electrophoresis of DNA through agarose. Under some conditions DNA molecules of the same length but with different linking numbers can be made to separate from one another upon electrophoresis. The separation results from the fact that two molecules with different linking numbers will, on the average during the electro-phoresis, possess different degrees of supercoiling and consequently different compactness. Those molecules that are more greatly super-coiled during the electrophoresis will migrate more rapidly. Not only can agarose gels be used for quantitating species with different numbers of superhelical turns, but any particular species can be extracted out of the gel and used in subsequent experiments.

The agarose gels show an interesting result. If DNA is ligated to form covalently closed circles and then subjected to electrophoresis under conditions that separate superhelical forms, it is found that not all of the DNA molecules possess the same linking number. There is a distri-bution centered about the linking number corresponding to zero super-helical turns, Lk₀. This is to be expected because the DNA molecules in solution are constantly in motion, and a molecule can be ligated into a covalently closed circle at an instant when it possesses a linking number unequal to Lk₀. These molecules are frozen in a slightly higher average energy state than those with no superhelical turns. Their exact energy depends on the twisting spring constant of DNA. The stiffer the DNA, the smaller the fraction of molecules that will possess any superhelical turns at the time of sealing. Quantitation of the DNA molecules in the bands possessing different numbers of superhelical turns permits evalu-ation via statistical mechanics of the twisting spring constant of DNA.

The ability to measure accurately the number of superhelical turns in DNA allows a determination of the amount of winding or unwinding produced by the binding of molecules. For example, unwinding meas - urements first indicated that RNA polymerase melts about 8 bases of DNA when it binds tightly to lambda DNA. Later, more precise meas-urements have shown that the unwinding is closer to 15 base pairs. This unwinding was shown directly by binding RNA polymerase to nicked circular DNA and then sealing with ligase to form covalently closed circles, removing the RNA polymerase, and measuring the number of superhelical turns in the DNA. The first measurements were done by accurately comparing the sedimentation velocity of the DNA sealed in the presence and in the absence of RNA polymerase. Later experiments have used a better DNA substrate and have used gel electrophoresis.

Another way to measure the winding produced by binding of a molecule to DNA is to measure the affinity of a molecule for DNA samples containing different numbers of superhelical turns. This method is based on the fact that a protein which introduces negative superhelical turns as it binds to DNA will bind much more tightly to a DNA molecule already containing negative superhelical turns. From the thermodynamics of the situation, this type of approach is very sensitive.