Vanderbilt Institute of Chemical Biology



Discovery at the VICB







Targeting DNA Repair with DSS1


By: Carol A. Rouzer, VICB Communications
Published:  July 21, 2015



DSS1 targets BRCA2 to RPA at sites of DNA damage, enabling it to facilitate the hand-off to RAD51, an important step in double strand break repair.


DNA double strand breaks (DSBs) are a particularly dangerous form of damage that can result from certain types of structural lesions or stalled replication forks. Efficient, error-free repair of DSBs is required for cells to maintain genomic integrity during DNA synthesis. Homologous recombination (HR, Figure 1) is a highly conserved pathway that leads to error-free DSB repair. During HR, nucleases cleave off a number of nucleotides from one strand of the DNA at each of the broken ends. The remaining single-stranded DNA (ssDNA) then rapidly becomes the binding site for a number of repair proteins. First in line is replication protein A (RPA), a protein that is well known for its ability to bind to and protect ssDNA. Eventually, however, RPA must be displaced by RAD51, a recombinase that forms a helical filament around the ssDNA. RAD 51 then searches for the complementary region on the homologous, undamaged chromosome, and catalyzes joint formation, the annealing of the single strand from the damaged chromosome with its complement on the homologous chromosome. This sets the stage for error-free DNA synthesis that fills the gap in the damaged chromosome and ultimately allows for complete repair. Numerous proteins are involved in RAD51’s ability to displace RPA on ssDNA. One of these is BRCA2, although how BRCA2 is involved in this process is not clear. Now, Vanderbilt Institute of Chemical Biology member Walter Chazin and his collaborator Patrick Sung (Yale University) define the role of DSS1 in BRCA2-mediated RPA displacement by RPA51 and its importance to HR. [W. Zhao, et al. (2015) Mol. Cell, published online July 1, DOI:10.1016/j.molcel.2015.05.032.



Figure 1. Diagrammatic representation of homologous recombination. (a) DNA damage leads to a double-strand break (DSB) in one of a pair of homologous chromosomes. (b) Activation of the ATM kinase follows, which leads to subsequent activation of multiple DNA repair proteins. (c) A nuclease removes a portion of one strand of the broken duplex, yielding a single-stranded region. RPA binds to this region, followed by RAD51 with the help of RAD52, and other proteins, including BRCA2. (d) RAD51 forms a helical filament on the single-stranded DNA, and then (e) searches for the complementary region on the undamaged chromosome. The homologous regions of the two chromosomes are then annealed in the joint formation step, and (f) DNA synthesis fills in the gap in the damaged chromosome. (g) Finally, the repaired strands of the damaged chromosome are annealed and ligated, resulting in a complete, error-free repair. Figure reproduced by permission from Macmillan Publishers, Ltd. from D. C. van Gent, et al., (2001) Nat. Rev. Genetics, 2, 196. Copyright 2001.


BRCA2 plays an important role in the maintenance of genome integrity. People carrying BRCA2 mutations display increased sensitivity to genotoxic agents and replication stress. BRCA2 mutations are also associated with breast, ovarian, and other forms of cancer. Past research has shown that BRCA2 interacts with RAD51 through its eight breast cancer (BRC) repeat domains. BRCA2 also contains a DNA binding domain (DBD), comprising three oligonucleotide binding (OB) folds (Figure 2). The ability of BRCA2 to enhance the assembly of RAD51 on ssDNA in the presence of RPA is well documented; however, BRCA2 does not directly bind to RPA. This led the research team to hypothesize that a BRCA2 binding partner acts as an intermediary between BRCA2 and RPA.



Figure 2. Previously proposed model for BRCA2-mediated binding of RAD51 to ssDNA. RAD51 binds to BRCA2’s BRC repeats, shown as purple spikes interdigitating between the RAD51 monomers. BRCA2 then binds to the ssDNA via its OB folds, guiding the RAD51 monomers onto the ssDNA as it displaces the previously bound RPA. Figure reproduced by permission from Macmillan Publishers, Ltd. from S. N. Powell and L. A. Kachnic, (2003) Oncogene, 22, 5784. Copyright 2003.



DSS1 is a small, highly acidic protein that has been associated with a variety of cancers. Known for its important role in DSB and replication fork repair, DSS1 associates with the OB1 fold and an adjoining region of BRCA2, suggesting that it might be the binding partner that the researchers had proposed. To test this hypothesis, the first step was to find out if DSS1 alters the interaction of BRCA2 with ssDNA in the presence of RPA. An important advance toward achieving this goal was the successful expression and purification of BRCA2, DSS1, miBRCA2 (containing BRC domain 4, the DBD, and the C-terminus of BRCA2), and complexes of BRCA2 and miBRCA2 with DSS1. Using two in vitro assays, the researchers showed that RPA suppressed the interaction of RAD51 with ssDNA. This suppression was reversed by both miBRCA2 and an miBRCA2-DSS1 complex, but the miBRCA2-DSS1 complex was much more potent than miBRCA2 alone. The investigators were interested to find that addition of pure DSS1 and miBRCA2 individually to their assay mixtures did not result in the same increased potency observed with the preformed miBRCA2-DSS1 complex. They also found that miBRCA2-DSS1 did not show higher potency than miBRCA2 in facilitating RAD51 binding to ssDNA in the presence of the ssDNA binding protein SSB, suggesting specificity for RPA.


Prior work had shown that DSS1 does not bind to ssDNA, double-stranded DNA (dsDNA), or RAD51. However, an affinity pulldown assay demonstrated that DSS1 does bind directly to RPA, and that it weakens RPA’s interaction with ssDNA. Furthermore, the miBRCA2-DSS1 complex, but not miBRCA2 alone, binds to RPA. Consistent with their functional studies, the investigators also showed that DSS1 does not bind to SSB.


RPA has three subunits, RPA70, RPA32, and RPA14 (Figure 3). Of the three, RPA70 is the most complex, containing three DBDs designated RPA70A, B, and C, and a protein interaction domain, RPA70N. RPA32 also contains a DNA binding domain (RPA32D) as well as a protein interaction domain (RPA32C) and a disordered regulatory domain (RPA32N). Affinity pulldown assays showed that DSS1 interacts in multiple ways with all of the major domains of RPA70, but not with RPA32 or RPA14. Isothermal titration calorimetry confirmed these findings and demonstrated a 1-to-1 interaction between DSS1 and RPA. Using 15N-1H-HSQC (heteronuclear single quantum coherence) NMR spectroscopy, the investigators were able to identify key residues in RPA70 that are involved in its interaction with DSS1. They found that DSS1 interacts with basic clefts in RPA that are used for binding to ssDNA and that ssDNA competes with DSS1 for those RPA binding sites.




Figure 3. (A) Domain structure of the three subunits of RPA. (B) RPA14, the RPA32D domain, and the RPA70C domain form a trimer in solution. The remaining domains are freely mobile on flexible tethers until the protein interacts with specific binding partners.



Having established that DSS1 exploits a DNA binding site to interact with RPA, the investigators started a search for the region of DSS1 that is responsible for this interaction. They identified a segment in the middle of DSS1 that comprises mostly acidic residues. Hypothesizing that this region might mimic the acidic ssDNA structure, they mutated eight of the residues to alanine, and tested the ability of the resulting DSS18A mutant protein to substitute for DSS1 in their various assays. They found that, although DSS18A exhibited normal binding to BRCA2 or miBRCA2, its ability to interact with RPA was markedly impaired. Furthermore, the miBRCA2-DSS18A complex showed no increased potency over miBRCA2 alone in its ability to facilitate binding of RAD51 to ssDNA in the presence of RPA.


Having firmly established that DSS1 mediates an interaction between miBRCA2 and RPA, the investigators went on to confirm that these results also applied to full length BRCA2. They then turned their attention to the role of DSS1 in HR in intact cells. They discovered that siRNA-mediated knockdown of DSS1 suppressed HR in U2OS cells, and that HR could be restored by expressing siRNA-resistant wild-type DSS1, but not DSS18A in the cells. Then, using HeLa cells, they demonstrated that X-ray exposure resulted in the formation of RAD51 foci in the nuclei of the cells, an indication of ongoing HR (Figure 4). Again, siRNA-mediated knockdown of DSS1 suppressed the formation of these foci, while the effects of the knockdown were abrogated by expression of wild-type DSS1, but not DSS18A. Similarly, knockdown of DSS1 increased the sensitivity of the cells to the genotoxicants mitomycin C and Olaparib, and these effects could also be counteracted by expression of wild-type DSS1, but not the DSS18A mutant.




Figure 4. Formation of RAD51 foci in response to DNA damage. The cells on the right (b) were exposed to radiation while the cells on the left (a) served as controls. RAD51 accumulates on the DNA at the sites of damage, forming foci that can be seen in the nuclei of the radiation-treated cells by immunofluorescence microscopy. Figure reproduced by permission from Macmillan Publishers, Ltd. from D. C. van Gent, et al., (2001) Nat. Rev. Genetics, 2, 196. Copyright 2001.



The Chazin and Sung team conclude that DSS1 targets BRCA2 to RPA and then serves as an ssDNA mimic to suppress the binding of RPA to ssDNA. This allows BRCA2 to guide RAD51 onto the ssDNA strand. These findings provide the answers to a long-standing question about the mechanism of HR, i.e., how RPA releases its tight grip so that Rad51 can be loaded and perform the vital homology search. Overall, these new results enhance our understanding of a critical pathway required to maintain a healthy genome in the face of the numerous challenges from the constant barrage of DNA damage and replicative stress each day.









The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center