HENRY M. SOBELL

EXPERIMENTAL PREDICTIONS

My theory makes a number of testable experimental predictions. The existence of beta-DNA, "pinned" by intercalators, can readily be checked by X-ray crystallography. Supporting evidence for the existence of premeltons in DNA can be provided using standard techniques available in molecular biology and physical chemistry.

  1. X-ray crystallographic experiments

    The theory predicts that simple intercalators (i.e., ethidium, acridine orange, ellipticine, terpyridine platinum and tetramethyl-N-methylphenantrolinium) bind to the high-energy beta-DNA form, while partial intercalators (i.e., irehdiamine and dipyrandium) bind to the low-energy beta-DNA form. Neighbor-exclusion binding reflects the presence of beta-DNA in both types of complexes.

    1. For these reasons, a search should be initiated for crystals that contain both simple and partial intercalators complexed to a series of self-complementary DNA oligonucleotides. If successful, the structures of these complexes can then be determined by X-ray crystallography.

      Ideally, these studies can be carried out as a matrix of interactions having two variables. Self-complementary oligonucleotides with constant length but containing differing nucleotide base sequences comprise the first variable. Simple intercalators consisting of different heterocyclic ring systems, as well as partial intercalators (these contain a variety of steroidal diamines having differing stereochemistries), constitute the second variable.

      Additional studies can be carried out to establish the structures of the oligonucleotides in their uncomplexed form.

      In this exhaustive study, the theory predicts each oligomer duplex to exist in the high-energy beta-DNA form when "pinned" by simple intercalators, and in the low-energy beta-DNA form when "pinned" by partial intercalators.

      Note: Propidium contains two positive charges, while the self-complementary DNA-like hexanucleotide duplex, TACGTA, contains ten negative charges. Intercalation by propidium is known to favor pyrimidine-purine sequences (refer to the earlier discussion regarding this), and for this reason, three propidium molecules are expected to intercalate between TA, CG and TA sequences. Two additional propidium molecules can then stack above and below this intercalated hexanucleotide to maintain overall charge neutrality in forming this solid state complex. A similar structural motif has been observed in crystalline complexes containing simple intercalators complexed to a variety of self-complementary DNA and RNA dinucleoside monophosphates (again, the reader should refer to the earlier discussion that describes these structures). For these reasons, this may be an excellent candidate to begin these studies.

      Note: Both irehdiamine and dipyrandium also contain two positive charges, and, for this reason, they may be appropriate candidates for similar studies.

    2. Parallel studies should be carried out with RNA oligomers having similar sequences. The presence of the 2' hydroxyl group in ribose sugar residues can be readily detected in difference Fourier electron density maps and, therefore, provide a convenient "signature" to pinpoint sugar puckering. Again, such studies should present a detailed comparison between conformations of RNA duplexes in both the complexed and uncomplexed forms.

    Actinomycin is proposed to bind to beta-DNA found within the boundaries connecting double-stranded B-DNA with single-stranded DNA in the transcription complex. This immobilizes (i.e., "pins") the complex, interfering with the elongation of growing RNA chains. In extremely active genes, RNA polymerases lie in a close-packed arrangement along DNA. Interference with the movement of one polymerase by actinomycin is expected to affect the movement of other polymerases. This can explain why nucleolar RNA synthesis is so sensitive to the presence of actinomycin D.

    1. The theory proposes actinomycin to be an example of a complex intercalator that binds to beta-DNA, a prediction that can readily be tested by X-ray crystallography.

      For example, AGCAAGCT and AGCTTGCT are complementary single-stranded oligomer sequences that contain two actinomycin GC binding sites per duplex. The model predicts the phenoxazone ring system on each actinomycin molecule to intercalate into these GC sites, "pinning" the high-energy beta-structural element. The internal AA (and TT) site is "unpinned" and therefore remains as a low-energy beta-structural element. Due to the "end effect", it is not possible to predict the puckering of deoxyribose sugar residues in the terminal A (and T) regions.

  2. Molecular biology experiments.

    The theory predicts premeltons to undergo low amplitude breather motions within promoter regions, these serving as nucleation centers for site-specific DNA melting by the RNA polymerase enzyme (and, possibly, by other proteins that play a role in gene activation).

    As described previously, key evidence for the existence of premeltons in eukaryotic promoter regions has been provided by the discovery that the intercalator, 1, 10- phenanthroline-copper (I), cleaves DNA at hypersensitive sites in naked relaxed circular DNA molecules containing the histone gene cluster of D. melanogaster, many of these located at the 5' ends of genes. Remarkably, this small molecule mimics the larger micrococcal nuclease in recognizing these sites, cleaving them with equal frequencies (see earlier discussion). Pancreatic DNase (I) cleaves these same sites with similar frequencies, as well as other unrelated sites.

    Low amplitude breather motions within premeltons cause the central beta-structural element to alternate between its two different energy states. The highest energy state (i.e., the high-energy beta-structural element) is the substrate for both 1, 10- phenanthroline-copper (I) and the micrococcal nuclease, whereas the lowest energy state (i.e., the low-energy beta-structural element) is the substrate recognized by the pancreatic DNase (I).

    1. Experiments should be carried out with the histone gene cluster of D. melanogaster in which ethidium is used as a competitive inhibitor of the cleavage reaction mediated by 1, 10- phenanthroline-copper (I) and the micrococcal nuclease. Should nuclease hypersensitive sites disappear at elevated ethidium/DNA binding ratios, this would suggest the presence of tight binding sites for intercalation at these regions that inhibit catalysis. The presence of these tight binding sites is most easily understood as reflecting the presence of premeltons in these regions.

      Intercalation by ethidium into the centers of premeltons is also expected to inhibit pancreatic DNase cleavage at these sites. This is because the act of "pinning" a high-energy beta-structural element by intercalation interferes with the appearance of a low-energy beta-structural element.

      The theory further predicts that sites uniquely cleaved by the pancreatic DNase should remain hypersensitive in the presence of ethidium. Premeltons containing lower energies are expected to undergo smaller breather motions that may not be able to open the central beta-structural element wide enough to be a substrate for 1, 10- phenanthroline copper (I) or the micrococcal nuclease. Premeltons such as these however can remain as substrates for the pancreatic DNase (I), even in the presence of ethidium.

    2. Competition experiments should also be carried out with irehdiamine (and other related steroidal diamines) using this same experimental system. Again, the theory predicts nuclease hypersensitive sites recognized by all three agents disappear with increasing concentrations of irehdiamine. This is because partial intercalation of irehdiamine into a low-energy beta-structural element interferes with the formation of a high-energy beta-structural element.

      The theory further predicts that sites uniquely cleaved by the pancreatic DNase should be protected from cleavage in the presence of irehdiamine. The reason for this follows from the discussion above.

    Micrococcal nuclease and pancreatic DNase (I) have been used to probe the structure of DNA in inactive chromatin. Micrococcal nuclease (as well as 1, 10- phenanthroline copper (I)) cuts DNA every 200 base pairs to generate the basic subunit structure of inactive chromatin, the nucleosome; while pancreatic DNase gives a more complex cleavage pattern, maximal cleavage occurring every 10, 11; 21, 22, etc. base pairs along DNA.

    These data are consistent with the presence of (static) high-energy beta- structural elements lying between nucleosomes, unwinding DNA every 200 base pairs to form the higher order solenoid structure in chromatin (i.e., the 300 Angstrom fiber), and by the statistical appearance of lower-energy beta- structural elements in nucleosomal DNA ("kinks", in the molecular biology sense) that transiently arise due to the presence of bending strain energy in its left-handed toroidal super-helical structure.

    1. Competition experiments should be carried out to examine whether increasing concentrations of ethidium inhibit micrococcal nuclease cleavage of DNA in inactive chromatin. If this were the case, this would suggest the presence of (static) high-energy beta-structural elements every 200 base pairs associated with the higher order solenoid structure in chromatin.

      The presence of ethidium would not be expected to interfere with the action of the pancreatic DNase, however, which cleaves DNA every 10, 11; 21, 22, etc. base pairs within the nucleosome. This is because the substrate for recognition is the low-energy beta-structural element, which arises transiently due to the presence of bending strain energy in the left-handed toroidal B-DNA super-helical structure.

    2. Similar competition experiments should be carried out to examine whether increasing concentrations of irehdiamine inhibit pancreatic DNase cleavage of DNA in inactive chromatin. If this were the case, this would suggest the presence of low-energy beta-structural elements (i.e., "kinks" in the molecular biology sense) that transiently arise in B-DNA due to the presence of bending strain energy in the left-handed toroidal super-helical structure within the nucleosome, as described above.

      The presence of irehdiamine would not be expected to interfere with the action of the micrococcal nuclease, however, which cleaves DNA every 200 base pairs between nucleosomes. This is because the high-energy beta-structural element is static (perhaps due to the presence of the H1 histone), a necessary structural feature to form the higher order solenoid structure in chromatin.

  3. Physical chemistry experiments.

    The appearance of premeltons in DNA is a physical property of its long chain polymer structure. In this sense, short oligonucleotide duplexes are not good models to understand DNA.

    How long must an oligonucleotide duplex be in order to "visualize" a premelton? In principle, the answer to this question can be approached by studying the appearance (i.e., emergence) of some known nuclease hypersensitive site as a function of the surrounding DNA chain length and sequence.

    Again, the histone gene cluster in D. melanogaster could be a good model system to study this. The general plan would be to synthesize a DNA region that contains (for example) the H-4 nuclease hypersensitive site located at the 5' end of the gene, and to then add varying lengths of the histone gene cluster sequence around this. The emergence of a premelton centered at this nuclease sensitive site could be detected by a gradual enhancement in the sensitivity for cleavage by 1, 10-phenanthroline copper (I). Ideally, two fragments would be generated by hydrolytic cleavage, and each could be detected by gel electrophoresis and then subsequently sequenced.

    The minimum size of a premelton is estimated to be between fifty to sixty base pairs. An overall length of one-hundred (or more) nucleotides may be necessary to begin to "visualize" a premelton, and to study its properties. Should this be possible, energetic factors governing the appearance of the premelton can then be investigated by systematically altering the nucleotide sequence at its most central region (i.e., the nuclease hypersensitive site) and its surrounding regions (i.e., kink and antikink regions).

    Typical experiments might include the deliberate insertion of a mismatched base pair in the central region, and investigating the sensitivity this region has to nucleases or chemical agents. The excision repair system can also be investigated to repair these mismatches, along with the formation and repair of pyrimidine dimers induced by ultraviolet radiation (i.e., the formation of the cyclobutane bridge connecting pyrimidine rings could be facilitated by the transient juxtaposition of adjacent pyrimidine rings while in the low-energy beta-structural element).

    Finally, for those interested to understand the mechanism of DNA melting, this line of research should offer rich rewards.