Jump to: Page Content, Section Navigation, Site Navigation, Site Search, Account Information, or Site Tools.
A Novel Configuration of DNA Induced by Homologous RecombinationTaro NishinakaThe Escherichia coli protein RecA is an essential protein during homologous recombination. Homologues of this protein are widely present in nature. They can be found in archaebacteria and in humans. RecA acts as a cofactor with ATP in exchange reactions of homologous DNA strands by searching for homologous base sequences between single-stranded DNA and double-stranded DNA and promoting base pairing. A proper understanding of this reaction mechanism requires knowledge of the various configurations of DNA during these reaction stages. If RecA is added to single-stranded DNA, the RecA binds around the single strands of DNA to form filamentous complexes. Observation by electron microscopy shows that the DNA in these complexes can be extended to about 1.5 times the comparable length of B-DNA, with approximately 19 bases per turn observed on unwinding (B-DNA has about 10.5 base pairs per turn). However, there have been no reports of X-ray crystallographic and nuclear magnetic resonance (NMR) structural analyses of these DNA complexes, and information at the atomic coordinate level is not yet available. NMR analysis of macromolecules, such as complexes between RecA and long strands of DNA, is extremely difficult because of signal broadening and indistinct resolution of the resonance lines. In fact, no successful NMR studies of RecA or corresponding complexes with DNA have been reported. My doctoral research study posed the question of whether information about the structure of single-stranded DNA bound to RecA could be obtained by TRNOE (transferred nuclear Overhauser effect) spectroscopy in the presence of rapid binding-dissociation chemical exchanges of DNA with RecA using short DNA oligomers. After several years investigating optimal measurement parameters and improving the measurement technique, I was actually able to determine the structure of single-stranded DNA bound to RecA. TRNOE spectroscopy is one general type of NMR measurement that can analyze the structure of small molecules bound to macromolecules when binding and dissociation of small molecules with these macromolecules occurs at rapid exchange rates. Figure 1 shows a structural model of DNA bound to RecA based on TRNOE spectroscopic data (1). The most prominent structural characteristic is the stacking configuration. In contrast to the van der Waals forces acting between bases in typical DNA, the bases are further separated in the RecA-bound form of DNA. The deoxyribose methylene (C2'-H2'-H2") sites are positioned above the subsequent residue base sites. This deoxyribose-base stacking is thought to contribute to stabilization of the entire structure by means of hydrophobic interactions. If the helical axis toward the RecA filament is positioned almost perpendicular to the plane of the bases, the distance between bases is about 5 Å. This is equivalent to about 1.5 times the corresponding distance in B-DNA. This experiment is the first to clearly demonstrate at the atomic coordinate level the specific DNA configuration formed during homologous recombination. Figure 1. Structual Model of Single-stranded DNA bound to RecA protein.
After determining the structure of single-stranded DNA bound to RecA using NMR, I next attempted to construct a structural model of double-stranded DNA bound to RecA. In the presence of ATP, RecA was also able to bind to double-stranded DNA, forming a filamentous complex having a pitch of approximately 95 Å. As in the case of single-stranded DNA, the double-stranded DNA contained within the RecA filament also could be extended to about 1.5 times the length of typical DNA, with about 18.6 base pairs per turn observed on unwinding. I first constructed a model of the double-stranded DNA structure on the basis of the hypothesis that the DNA sugar puckering was of the S-type (C2' endo). This resulted in a double-stranded DNA structure with a pitch (12.5 base pairs per turn) of approximately 64 Å (Fig. 2, center) (2). This corresponded to the pitch of a RecA ADP-type filament. This did not, however, result in a double-stranded DNA configuration with a pitch of 95 Å, which would correspond to the pitch of an ATP-type filament. However, double-stranded DNA with a pitch of approximately 95 Å was successfully obtained based on the hypothesis that sugar puckering was of the N-type (C3' endo) (Fig. 2, left) (2). Figure 2. Structual Model of Double-stranded DNA bound to RecA protein.
Although I was amazed at the thought of being able to create molecular structures with the respective pitches of RecA ATP-type and ADP-type filaments simply by changing the sugar puckering, the exact significance of these findings remained unclear. One spring evening I took a molecular structural model of DNA that I constructed myself into my hand and began to ponder the configuration of DNA within RecA filaments. I suddenly realized that altering the DNA sugar puckering would permit nearly horizontal rotation of the molecule while maintaining the extended configuration, and without any significant steric hindrance at the bases. (Fig. 3) (2). A change in DNA sugar puckering from N-type to S-type would result in base site rotation in the minor groove direction. If the respective bases of double-stranded DNA rotated by this mechanism, and if prior to this a single-stranded DNA existed in which new base pairs could be formed, then base pair exchange could occur (Fig. 4). These base rotations due to alterations in sugar puckering probably play an extremely important role in homologous recombination reactions in which homologous base sequences are searched for, with subsequent strand exchange. In fact, NMR relaxation matrix analysis suggests an equilibrium between N-type and S-type sugar puckering in single-stranded DNA molecules bound to RecA. Figure 3. Base Rotation by Interconversion of Sugar Puckering. Figure 4. Structural Model of Triple-stranded DNA for Homology Search and Strand Exchange.
The principle objective of my doctoral thesis was to demonstrate the possibility of an important association of the chemical structure of DNA with a homologous recombination reaction, which is one well-known type of basic biological phenomenon. The results of my analyses using NMR show that it is primarily the intrinsic chemical structural characteristics of the DNA that maintain the extended stereospecific structure of the molecule rather than the intercalation of the amino acid residues of RecA between bases of DNA. This configuration is additionally characterized by relatively free rotation of the bases. This degree of flexibility is probably a factor in enabling base pair exchange dynamic reactions to occur between single-stranded DNA and double-stranded DNA. Biochemical studies performed to date indicate that RNA has a low affinity for RecA. Because the H2" at the C2'-H2'-H2" site within the stacking configuration is replaced by an OH radical in RNA, steric hindrance with subsequent bases occurs in the RNA molecule. Therefore, this type of structure is thought to be unstable in RNA. Almost all living organisms use DNA as the vehicle for transmitting genetic information. There are very few examples of organisms such as the RNA viruses that use RNA as their genome. The reason is that the H2" of DNA, which is absent in RNA, results in chemical stability of the DNA. This explanation is widely accepted as being the reason why the DNA molecule represents a more advantageous site for preserving genetic information. The results of my NMR analyses suggest that the residue interactions between the deoxyribose C2'-H2'-H2" sites and the subsequent residue base sites are a primary factor in the extended structure of the DNA, and that this plays a key role in efficiently allowing homologous recombination to occur. In addition to the role of DNA in preserving genetic information, the H2" of DNA may also play an important role in reorganization and diversity of genes (namely, homologous recombination). I believe that this stereochemical structural advantage of genetic substances may have become necessary during the process of evolution. Although absent in RNA, this seems to be an intrinsic characteristic of the DNA molecule.
References
|
Science. ISSN 0036-8075 (print), 1095-9203 (online)