Replication protein A interactions with DNA. III. Molecular basis of recognition of damaged DNA

Human replication protein A (RPA) is a heterotrimeric single-stranded DNA-binding protein (subunits of 70, 32, and 14 kDa) that is required for cellular DNA metabolism. RPA has been reported to interact specifically with damaged double-stranded DNA and to participate in multiple steps of nucleotide excision repair (NER) including the damage recognition step. We have examined the mechanism of RPA binding to both single-stranded and double-stranded DNA (ssDNA and dsDNA, respectively) containing damage. We show that the affinity of RPA for damaged dsDNA correlated with disruption of the double helix by the damaged bases and required RPAs ssDNA-binding activity. We conclude that RPA is recognizing single-stranded character caused by the damaged nucleotides. We also show that RPA binds specifically to damaged ssDNA. The specificity of binding varies with the type of damage with RPA having up to a 60-fold preference for a pyrimidine(6-4)pyrimidone photoproduct. We show that this specific binding was absolutely dependent on the zinc-finger domain in the C-terminus of the 70-kDa subunit. The affinity of RPA for damaged ssDNA was 5 orders of magnitude higher than that of the damage recognition protein XPA (xeroderma pigmentosum group A protein). These findings suggest that RPA probably binds to both damaged and undamaged strands in the NER excision complex. RPA binding may be important for efficient excision of damaged DNA in NER.     

Replication protein A interactions with DNA. 2. Characterization of double-stranded DNA-binding/helix-destabilization activities and the role of the zinc-finger domain in DNA interactions

Human replication protein A (RPA) is a heterotrimeric single-stranded DNA-binding protein that is composed of subunits of 70, 32, and 14 kDa. RPA is required for multiple processes in cellular DNA metabolism. RPA has been reported to (1) bind with high affinity to single-stranded DNA (ssDNA), (2) bind specifically to certain double-stranded DNA (dsDNA) sequences, and (3) have DNA helix-destabilizing ("unwinding") activity. We have characterized both dsDNA binding and helix destabilization. The affinity of RPA for dsDNA was lower than that of ssDNA and precisely correlated with the melting temperature of the DNA fragment. The rates of helix destabilization and dsDNA binding were similar, and both were slow relative to the rate of binding ssDNA. We have previously mapped the regions required for ssDNA binding [Walther et al. (1999) Biochemistry 38, 3963-3973]. Here, we show that both helix-destabilization and dsDNA-binding activities map to the central DNA-binding domain of the 70-kDa subunit and that other domains of RPA are needed for optimal activity. We conclude that all types of RPA binding are manifestations of RPA ssDNA-binding activity and that dsDNA binding occurs when RPA destabilizes a region of dsDNA and binds to the resulting ssDNA. The 70-kDa subunit of all RPA homologues contains a highly conserved putative (C-X2-C-X13-C-X2-C) zinc finger. This motif directly interacts with DNA and contributes to dsDNA-binding/unwinding activity. Evidence is presented that a metal ion is required for the function of the zinc-finger motif.     

Structure and function of a novel endonuclease acting on branched DNA substrates

We show that Pyrococcus abyssi PAB2263 (dubbed NucS (nuc lease for s s DNA) is a novel archaeal endonuclease that interacts with the replication clamp PCNA. Structural determination of P. abyssi NucS revealed a two‐domain dumbbell‐like structure that in overall does not resemble any known protein structure. Biochemical and structural studies indicate that NucS orthologues use a non‐catalytic ssDNA‐binding domain to regulate the cleavage activity at another site, thus resulting into the specific cleavage at double‐stranded DNA (dsDNA)/ssDNA junctions on branched DNA substrates. Both 3′ and 5′ extremities of the ssDNA can be cleaved at the nuclease channel that is too narrow to accommodate duplex DNA. Altogether, our data suggest that NucS proteins constitute a new family of structure‐specific DNA endonucleases that are widely distributed in archaea and in bacteria, including Mycobacterium tuberculosis.     

A toolbox for generating single‐stranded DNA in optical tweezers experiments

Essential genomic transactions such as DNA‐damage repair and DNA replication take place on single‐stranded DNA (ssDNA) or require specific single‐stranded/double‐stranded DNA (ssDNA/dsDNA) junctions (SDSJ). A significant challenge in single‐molecule studies of DNA–protein interactions using optical trapping is the design and generation of appropriate DNA templates. In contrast to dsDNA, only a limited toolbox is available for the generation of ssDNA constructs for optical tweezers experiments. Here, we present several kinds of DNA templates suitable for single‐molecule experiments requiring segments of ssDNA of several kilobases in length. These different biotinylated dsDNA templates can be tethered between optically trapped microspheres and can, by the subsequent use of force‐induced DNA melting, be converted into partial or complete ssDNA molecules. We systematically investigated the time scale and efficiency of force‐induced melting at different ionic strengths for DNA molecules of different sequences and lengths. Furthermore, we quantified the impact of microspheres of different sizes on the lifetime of ssDNA tethers in optical tweezers experiments. Together, these experiments provide deeper insights into the variables that impact the production of ssDNA for single molecules studies and represent a starting point for further optimization of DNA templates that permit the investigation of protein binding and kinetics on ssDNA. © 2013 Wiley Periodicals, Inc. Biopolymers 99:611–620, 2013.     

Counterintuitive DNA Sequence Dependence in Supercoiling-Induced DNA Melting

The metabolism of DNA in cells relies on the balance between hybridized double-stranded DNA (dsDNA) and local de-hybridized regions of ssDNA that provide access to binding proteins. Traditional melting experiments, in which short pieces of dsDNA are heated up until the point of melting into ssDNA, have determined that AT-rich sequences have a lower binding energy than GC-rich sequences. In cells, however, the double-stranded backbone of DNA is destabilized by negative supercoiling, and not by temperature. To investigate what the effect of GC content is on DNA melting induced by negative supercoiling, we studied DNA molecules with a GC content ranging from 38% to 77%, using single-molecule magnetic tweezer measurements in which the length of a single DNA molecule is measured as a function of applied stretching force and supercoiling density. At low force (<0.5pN), supercoiling results into twisting of the dsDNA backbone and loop formation (plectonemes), without inducing any DNA melting. This process was not influenced by the DNA sequence. When negative supercoiling is introduced at increasing force, local melting of DNA is introduced. We measured for the different DNA molecules a characteristic force F _char, at which negative supercoiling induces local melting of the dsDNA. Surprisingly, GC-rich sequences melt at lower forces than AT-rich sequences: F _char = 0.56pN for 77% GC but 0.73pN for 38% GC. An explanation for this counterintuitive effect is provided by the realization that supercoiling densities of a few percent only induce melting of a few percent of the base pairs. As a consequence, denaturation bubbles occur in local AT-rich regions and the sequence-dependent effect arises from an increased DNA bending/torsional energy associated with the plectonemes. This new insight indicates that an increased GC-content adjacent to AT-rich DNA regions will enhance local opening of the double-stranded DNA helix.     

Adjacent residues in the E1 initiator beta-hairpin define different roles of the beta-hairpin in Ori melting, helicase loading, and helicase activity

We have analyzed two residues in the helicase domain of the E1 initiator protein. These residues are part of a highly conserved structural motif, the beta-hairpin, which is present in the helicase domain of all papovavirus initiator proteins. These proteins are unique in their ability to transition from local template melting activity to unwinding. We demonstrate that the beta-hairpin has two functions. First, it is the tool used by the E1 double trimer (DT) to pry open and melt double-stranded DNA. Second, it is required for the unwinding activity of the hexameric E1 helicase. The fact that the same structural element, but not the same residues, contacts both dsDNA in the DT for melting and ssDNA in the double hexamer (DH) for helicase activity provides a link between local origin melting and DNA helicase activity and suggests how the transition between these two states comes about.     

Mechanical measurement of single-molecule binding rates: kinetics of DNA helix-destabilization by T4 gene 32 protein

Bacteriophage T4 gene 32 protein (gp32) is a single-stranded DNA (ssDNA) binding protein, and is essential for DNA replication, recombination and repair. While gp32 binds preferentially and cooperatively to ssDNA, it has not been observed to lower the thermal melting temperature of natural double-stranded DNA (dsDNA). However, in single-molecule stretching experiments, gp32 significantly destabilizes lambda DNA. In this study, we develop a theory of the effect of the protein on single dsDNA stretching curves, and apply it to the measured dependence of the DNA overstretching force on pulling rate in the presence of the full-length and two truncated forms of the protein. This allows us to calculate the rate of cooperative growth of single clusters of protein along ssDNA that are formed as the dsDNA molecule is stretched, as well as determine the site size of the protein binding to ssDNA. The rate of cooperative binding (ka) of both gp32 and of its proteolytic fragment *I (which lacks 48 residues from the C terminus) varies non-linearly with protein concentration, and appears to exceed the diffusion limit. We develop a model of protein association with the ends of growing clusters of cooperatively bound protein enhanced by 1-D diffusion along dsDNA, under the condition of protein excess. Upon globally fitting ka versus protein concentration, we determine the binding site size and the non-cooperative binding constants to dsDNA for gp32 and I. Our experiment mimics the growth of clusters of gp32 that likely exist at the DNA replication fork in vivo, and explains the origin of the "kinetic block" to dsDNA melting by gene 32 protein observed in thermal melting experiments.