Dynamic Assembly Of Reca Filaments On Single Molecules Of Ssb Coated Ssdna

In all organisms, homologous recombination is essential for the efficient and high fidelity repair of DNA lesions. Defects in homologous recombination result in genomic instability, which manifests in humans as a genetic predisposition to cancer. Central to the process of homologous recombination is the strand exchange activity of RecA (Escherichia coli), which forms a filament on single-stranded DNA (ssDNA) and aligns a broken chromosome with an intact one (a process called synapsis). Filament formation is inhibited by the diffusion-limited association of ssDNA with the high-affinity single-stranded DNA binding protein, SSB. A class of positive regulators called mediators facilitates filament formation by alleviating this biochemical inhibition. These mediators include RecFOR and RecBCD (Escherichia coli), Rad52 (Saccharomyces cerevisiae and Homo sapiens), and BRCA2 (Homo sapiens). Mediators of homologous recombination might regulate filament formation by accelerating either the frequency of nucleation, the rate of filament growth, or both; however, these kinetic parameters are indistinguishable in traditional biochemical (or bulk phase) experiments. Therefore, in order to fully understand the kinetic mechanism by which RecA filament assembly is regulated by mediator proteins, it is necessary to first determine the kinetic mechanism of spontaneous filament formation. To this end, I have developed several biochemical reagents and techniques that have facilitated the visualization of single molecules of ssDNA and used them to directly measure nucleation and growth of RecA on SSB-coated ssDNA. The manipulation of ssDNA in single-molecule assays is difficult due to poor methods of visualization as well as the complex polymer dynamics of ssDNA that result in self-condensation and the formation of secondary structure. To address these problems, fluorescent biosensors for ssDNA were created from both SSB and RPA using cysteine and N-terminal amine chemical modifications, respectively, with various fluorophores. Here, I characterized the ssDNA-binding properties of these biosensors using direct binding titrations and competition experiments. Each of the SSB-derived biosensors binds to ssDNA with nanomolar affinity, but are displaced from ssDNA by RecA between 17- and 27-fold faster than unmodified SSB. The affinity of the RPA-derived biosensors remains unaltered. Some of the properties of these biosensors, specifically the fluorescence enhancement upon binding to ssDNA and the ability to stimulate RecA- or Rad51-mediated strand exchange, vary in a fluorophore-dependent manner and were characterized and described. Importantly, each of these biosensors are suitable for the visualization of single molecules of denatured [lambda] phage DNA either tethered to a glass surface using total internal reflection fluorescence (TIRF) microscopy or tethered to a bead immobilized in an optical trap and visualized using epifluorescence microscopy. These biosensors are powerful tools in the manipulation and visualization of ssDNA in single-molecule assays. Using TIRF microscopy, I have directly visualized the assembly of RecA filaments on single molecules of SSB-coated ssDNA, revealing that a dimer of RecA is required for nucleation, followed by bidirectional growth of the filament through monomer addition, albeit with faster assembly in the 5'[right arrow]3' direction. Nucleation, but not growth, is modulated by nucleotide and magnesium ion cofactors, where the rate of nucleation in the presence of ATP[gamma]S and dATP are 10-fold and 5-fold greater than ATP. Nucleation and growth are both severely repressed by increasing pH and almost completely inhibited above pH 7.5, the intracellular pH of E. coli. The results presented here indicate that the rate of spontaneous nucleation on SSB-coated ssDNA requires two simultaneous events: 1) collision (and capture) of a RecA dimer with the ssDNA site exposed as 2) two protomers of SSB microscopically diffuse, or slide, away from each other, an event would be expected to be rare on a crowded (i.e., nearly saturated) ssDNA lattice. Finally, during direct visualization experiments, I observed a salt-induced ultra-structural condensation of single molecules of SSB-coated ssDNA corresponding to known `binding-mode' transitions. Unexpectedly, the observed length changes exceeded my expectations based on known electron microscopy (EM) and atomic force microscopy (AFM) studies. The condensation could not be explained by protein dissociation as fluorescent intensity measurements of single-molecules remained unchanged during and after salt-induced condensation. Surprisingly, the condensation was reversible when the salt was removed from the microfluidic flow chamber in the absence of free protein, demonstrating that the condensation and de-condensation was due to intra-molecular redistribution of SSB along the ssDNA molecules. These observations prompted me to analyze the force-extension relationship (i.e., force spectroscopy) of single molecules of SSB-coated ssDNA using magnetic tweezers in a range of ionic conditions as well as in the presence of the mediator proteins, RecO and RecOR, neither of which alter the condensation of SSB-coated ssDNA.