An alternative form of replication protein a expressed in normal human tissues supports DNA repair

Abstract

Replication protein A (RPA) is a heterotrimeric protein complex required for a large number of DNA metabolic processes, including DNA replication and repair. An alternative form of RPA (aRPA) has been described in which the RPA2 subunit (the 32-kDa subunit of RPA and product of the RPA2 gene) of canonical RPA is replaced by a homologous subunit, RPA4. The normal function of aRPA is not known; however, previous studies have shown that it does not support DNA replication in vitro or S-phase progression in vivo. In this work, we show that the RPA4 gene is expressed in normal human tissues and that its expression is decreased in cancerous tissues. To determine whether aRPA plays a role in cellular physiology, we investigated its role in DNA repair. aRPA interacted with both Rad52 and Rad51 and stimulated Rad51 strand exchange. We also showed that, by using a reconstituted reaction, aRPA can support the dual incision/excision reaction of nucleotide excision repair. aRPA is less efficient in nucleotide excision repair than canonical RPA, showing reduced interactions with the repair factor XPA and no stimulation of XPF-ERCC1 endonuclease activity. In contrast, aRPA exhibits higher affinity for damaged DNA than canonical RPA, which may explain its ability to substitute for RPA in the excision step of nucleotide excision repair. Our findings provide the first direct evidence for the function of aRPA in human DNA metabolism and support a model for aRPA functioning in chromosome maintenance functions in nonproliferating cells.

FIGURE 1.

Quantitative PCR of RPA4 and RPA2 mRNAs from different tissues. Relative mRNA expression of RPA2 (black) and RPA4 (gray) was determined by the comparative Ct method. Errors bars indicate the average of three technical and two experimental replicates. A, cDNA was made from a panel of 20 normal human tissues (Ambion) and HeLa cells were either mock-transformed, and HeLa cells were transformed with GFP-RPA4 fusion protein under the control of the cytomegalovirus promoter.⁴ B, cDNA made from normal and tumor tissue samples (Ambion).

FIGURE 2.

aRPA interactions with Rad51 and Rad52 and stimulates strand exchange. Enzyme linked immunosorbent assay in which interactions were measured between different forms of RPA and either Rad52 (A) or Rad51 (B). Forms of RPA used: RPA (open diamonds), aRPA (black squares), and AA-His (gray triangles). Errors bars indicate the average of two or more independent replicates. BSA was used to determine nonspecific background in each assay; BSA values, generally <0.1 OD were subtracted. C, schematic of DNA strand exchange between circular single-stranded DNA (ss) and homologous linear double-stranded DNA (ds) to produce joint molecules (JM) and nicked circular DNA (NC). The asterisk shows the ³²P-label on each strand. D, DNA strand exchange assay where ϕX174 (+) strand was incubated with Rad51 followed by RPA/aRPA and ³²P-labeled XhoI linearized ϕX174 double-stranded DNA. Samples were deproteinized and reaction products were separated by electrophoresis through a 1.0% agarose gel. The positions of joint molecules, nicked circular DNA (NC DNA), double-stranded DNA (dsDNA), and displaced single-stranded DNA (ssDNA) are indicated.

FIGURE 3.

aRPA supports nucleotide excision repair. A, RPA and aRPA were separated by SDS-PAGE and visualized by Coomassie Blue staining. The position of each subunit is indicated along with approximate molecular mass (kDa). B, schematic of the nucleotide excision repair assay. An internally ³²P-labeled (circle) 140-bp duplex DNA substrate containing a single (6-4) photoproduct (triangle) is incubated with purified excision repair factors, which results in dual incisions and release of 24–32-nt-long damage-containing oligomers. C, damage-containing substrate was incubated with 60 nm XPA, 9 nm XPC, 4 nm XPF-ERCC1, 3 nm XPG, 12.5 nm THIIH supplemented with indicated amounts of RPA or aRPA where indicated. The location of the excision products is indicated to the right. The percent of substrate in each reaction undergoing excision is indicated. D, time course of reconstituted excision repair reactions containing optimal amounts of either aRPA (530 ng) or RPA (150 ng). E, quantification of excision repair assays. Results indicate the average and S.D. from three independent experiments.

FIGURE 4.

aRPA does not confer structure-specific endonuclease activity to XPF-ERCC1. A, schematic of XPF junction cutting assay. A 3′-labeled (circle) 70-mer DNA containing a central 20-nt unpaired region was incubated in reactions containing RPA or aRPA and XPF-ERCC1. Junction cutting activity by the XPF endonuclease was detected by electrophoresis of substrate on a denaturing polyacrylamide gel. B, sample incision assay containing XPF-ERCC1 (7.5 ng) and increasing amounts of either RPA or aRPA (0, 5, 10, 20, and 40 ng). C, quantification of data from panel B.

FIGURE 5.

aRPA has altered interactions with XPA. A, enzyme-linked immunosorbent assay with RPA (open diamonds) or aRPA (black squares) and XPA. Errors bars indicate the average of two or more independent replicates. BSA was used to determine nonspecific background in each assay; BSA values, generally <0.1 OD were subtracted. B, immunoblot analysis of RPA and aRPA showing that anti-RPA2 and anti-RPA4 antibodies specifically recognize the appropriate subunits in RPA and aRPA, respectively. C, MBP-tagged XPA immobilized on amylose resin was incubated with RPA or aRPA overnight at 4 °C and then analyzed by SDS-PAGE and Western blotting with a mixture of the indicated antibodies. The recovered MBP-XPA was stained with Coomassie Blue after SDS-PAGE. Input represents 10% (100 ng) of RPA and aRPA and 1 μg of MBP-XPA.

FIGURE 6.

Binding of XPA, RPA, and aRPA to damaged DNA. A, linearized, biotinylated pUC19 plasmid DNA treated with AAAF was immobilized on streptavidin-coated magnetic beads and then incubated for 1 h at room temperature with the indicated proteins. Beads were washed, and associated proteins were analyzed by SDS-PAGE and Western blotting (WB) with anti-MBP and anti-RPA1 antibodies. Reactions contained 25, 50, or 100 ng of immobilized DNA, 20 ng of XPA and 100 ng of RPA or aRPA. B, quantification of experiments performed as in A. Western blot signals were normalized to a standard amount of XPA, RPA, or aRPA in each experiment to determine the percentage of each protein associating with the DNA. Data represent averages and S.D. from six-to-eight independent experiments. C, unmodified or AAAF-treated pUC19 plasmid DNA (25, 50, or 100 ng) immobilized on magnetic beads was incubated with RPA or aRPA for 1 h at room temperature. The associated RPA and aRPA was detected by SDS-PAGE and Western blotting with an anti-RPA1 antibody. The graph represents the average amount of RPA1 observed on the 100 ng DNA sample in two independent experiments. Data were normalized to the RPA1 signal from aRPA in both experiments.