The Compatibility of Netropsin and Actinomycin Binding to Natural Deoxyribonucleic Acid*

The simultaneous binding of netropsin and actinomycin to four natural DNAs was studied to determine the influence of one ligand on the binding of the other. Actinomycin binds specifically to GC sites, whereas netropsin binds specifically to AT sites. Spectral titrations, thermal denaturation, and analytical buoyant density centrifugation were employed to measure the binding interference of these drugs. The binding of actinomycin to DNA was decreased by the presence of rietropsin. Increasing the GC content of the DNA resulted in a decreased effect of netropsin on actinomycin binding. Quantitative analysis of the binding parameters indicated that netropsin and actinomycin can bind in close proximity along the DNA chain. Supercoiled DNA gave the same result as linear DNA. These results imply that DNA can absorb alterations in conformation within a short distance. Prior studies have provided composite evaluations of the conformation of natural DNA. Studies such as x-ray diffraction (l-3), thermal denaturation (T,), and buoyant density have demonstrated the influence of environmental factors (such as cation or humidity) on the over-all characteristics, changes along the entire length, of DNA. However, prior from this laboratory have demonstrated that the properties and structure of a DNA are dictated

The simultaneous binding of netropsin and actinomycin to four natural DNAs was studied to determine the influence of one ligand on the binding of the other. Actinomycin binds specifically to GC sites, whereas netropsin binds specifically to AT sites. Spectral titrations, thermal denaturation, and analytical buoyant density centrifugation were employed to measure the binding interference of these drugs.
The binding of actinomycin to DNA was decreased by the presence of rietropsin.
Increasing the GC content of the DNA resulted in a decreased effect of netropsin on actinomycin binding. Quantitative analysis of the binding parameters indicated that netropsin and actinomycin can bind in close proximity along the DNA chain. Supercoiled DNA gave the same result as linear DNA. These results imply that DNA can absorb alterations in conformation within a short distance.
Prior studies have provided composite evaluations of the conformation of natural DNA. Studies such as x-ray diffraction (l-3), thermal denaturation (T,), and buoyant density have demonstrated the influence of environmental factors (such as cation or humidity) on the over-all characteristics, changes along the entire length, of DNA. However, prior studies from this laboratory (4,5) have demonstrated that the properties and structure of a DNA are dictated by the nucleotide sequence. This influence was demonstrated by a variety of physical, chemical, and enzymatic measurements on 14 helical biosynthetic DNAs with defined repeating nucleotide sequences. Hence, the over-all conformation of natural DNA is probably a composite of many types of DNA structures.
Our present goal is to determine the proximity of binding (total neighbor exclusion lengths) of actinomycin and netropsin in order to provide information on the range of influence of one ligand on the other. Also, these studies should give an indication of the conformational flexibility of DNA over the length of a few base pairs, as opposed to the more composite measurements.
These results may be relevant to the problem of packaging DNA in a phage head as well as to the conforma- D sites include three GC, GC sites and two each of AT,GC and GC,AT sites. One of the GC,GC sites has a 2-fold degeneracy and is counted twice. Previous studies (10) have indicated that the affinity of netropsin for three contiguous AT pairs was considerably greater than its affinity for one or two AT pairs. Therefore, all AT pairs were regarded as potential netropsin sites, but only those sites with two adjacent ATs interfered with actinomycin D-DNA complex formation. The goal of this analysis was to estimate the total neighbor exclusion lengths of actinomycin D and strongly bound netropsin. To correctly analyze the actinomycin D binding parameters with netropsin present, one must obtain equations from  We assume netropsin binds to one end of three AT pairs and excludes 1, base pairs. Actinomycin D is assumed to centrally bind to the sites previously described and exclude 1, sites (1, -1 base pairs are covered).
where R,,, is the saturating density of bound netropsin with three AT pairs per site. R,,, is PAT times the saturating

RF ON A (next section).
Although this analysis was theoretically nonrigorous, it indicated that netropsin and actinomycin D can bind DNA within a short distance from each other. The values of 1, = 5 and 1, = 3 correspond roughly to the van der Waals radii of the adsorbed 'molecules.
It is clear that 1, and 1, were not substantially larger than the value expected from studies of each drug separately (8, 10).
Thus linear DNA can accommodate actinomycin D and netropsin, which have different binding modes, within a few base pairs of each other. The DNA conformations of the actinomycin D and netropsin complexes can exist in close proximity.
An examination of the assumptions of the above analysis and their effect on the results will be made in the "Discussion."

Two-drug
Binding to Supercoiled DNA-The binding of actinomycin to supercoiled 4X174 RF1 DNA was examined in the presence and absence of netropsin. The purpose of these experiments was to determine if employing a supercoiled DNA affects the above results. For a linear DNA, conformational changes induced by ligand binding can be propagated to the free ends. Similar changes in a supercoiled DNA can change the tertiary structure. Since actinomycin D unwinds DNA and netropsin does not, the binding of actinomycin D to supercoiled DNA should initially be greater than for linear DNA. theoretical prediction of NB = 0.12 for lN + 1, = 8. Experiments performed with linear 4x174 DNA gave results similar to that shown in Fig. 3. These experiments indicated that covalent closure of the free ends of a linear DNA did not measurably affect the minimum distance between actinomycin and netropsin.
The effect of binding a saturating quantity of actinomycin to 4X174 DNA on the netropsin-DNA binding parameters also was examined.
This approach was technically more difficult than the one previously described because the absorbance change per bound netropsin was smaller than for actinomycin (10) Fig. 4 shows the T, change versus drug concentration for C. perfringens DNA. Titrating actinomycin to saturation increased the T,,, by 5.0". Netropsin yielded a T,,, increase of 14.6" at saturation.
When netropsin was titrated in the presence of saturating amounts of actinomycin D, the increase in T, was always less than the additive effect expected from the independent binding of each drug. At 0.2 mol both actinomycin D and netropsin/DNA nucleotide, AT,,, = 15.7', whereas AT,,, expected from independent binding was 19.6". Analytical Cs,SO, density gradient titrations of C. perfringens DNA with netropsin or actinomycin D, or both, were performed in the manner previously described (10). The nonadditivity of the density decrements (results not shown) were in complete agreement with the conclusion drawn from T, studies.
A similar result was obtained with E. coli DNA (results not shown). When this DNA was complexed with both saturating netropsin and actinomycin D, it showed a T, increase of 10". The expected elevation of T, from titrating both drugs separately was 12.5'. The results indicated that netropsin and actinomycin D binding interfere with each other. The larger interference for C. perfringens DNA was in agreement with the spectral titration results.

DISCUSSION
The objective of this study was to determine the proximity of binding (total neighbor exclusion lengths) of actinomycin D and netropsin in order to provide information on the range of influence of one ligand on the other. Actinomycin D specifically binds to GC sites (6-9) whereas netropsin specifically binds to AT sites (10). Also, an indication of the conformational flexibility of DNA might derive from this data. The spectral titration results yielded an estimate of eight base pairs for the shortest outside end to outside end distance between bound netropsin and actinomycin.
The van der Waals radius of bound netropsin has been estimated as three base pairs (lo), and actinomycin D appears to cover five to six base pairs (8, 15). This indicated that DNA can accommodate both of these ligands within a short distance despite the distortions in DNA structure created by these ligands. Since the evaluation of 1, + 1, critically depends on the assumptions underlying Equations 2, 3, and 4, we will review them here. It was assumed that all intercalative sites with a 3'-guanine were equivalent actinomycin D-binding sites. Although there is evidence that the base pairs adjacent to these sites influence actinomycin D binding (9), this effect appears to be small for natural DNAs. The 1, values listed by Miiller and Crothers (8) for DNAs with varying GC content all give the same approximate value of six. This would not be the case (10) if actinomycin D sites were strongly influenced by neighboring base pairs.
In the analysis, it was assumed that only the netropsin molecules which were bound to three contiguous AT pairs could eliminate actinomycin D sites. Previous studies (10) have indicated that netropsin can also bind to a site with only one or two consecutive AT pairs. The binding affinities, however, were weaker than for three AT pairs. Thus R,,, was, if anything, underestimated by this assumption. If the R,,, values were larger, then the theoretical estimate of 1, + 1, would be smaller (see Equation 4). A third assumption was that titrating the DNA with actinomycin D did not lower R,,,. This depends on the binding constants of the ligands for their respective sites, and their concentrations.
For (dG-dC), (dG-dC), the actinomycin D binding constant (9) (10). Increasing the salt concentration from 0.01 M to 0.1 M decreases actinomycin binding to DNA (8). Therefore, it was unlikely that R,., was substantially altered by adding small amounts of actinomycin D to DNA saturated with netropsin.