Limits...
Reconstitution of recombination-associated DNA synthesis with human proteins.

Sneeden JL, Grossi SM, Tappin I, Hurwitz J, Heyer WD - Nucleic Acids Res. (2013)

Bottom Line: The repair of DNA breaks by homologous recombination is a high-fidelity process, necessary for the maintenance of genome integrity.Thus, DNA synthesis associated with recombinational repair must be largely error-free.Translesion synthesis polymerase eta (η) also extends these substrates, albeit far less processively.

View Article: PubMed Central - PubMed

Affiliation: Department of Microbiology & Molecular Genetics, University of California, Davis, Davis, CA 95616-8665, USA.

ABSTRACT
The repair of DNA breaks by homologous recombination is a high-fidelity process, necessary for the maintenance of genome integrity. Thus, DNA synthesis associated with recombinational repair must be largely error-free. In this report, we show that human DNA polymerase delta (δ) is capable of robust DNA synthesis at RAD51-mediated recombination intermediates dependent on the processivity clamp PCNA. Translesion synthesis polymerase eta (η) also extends these substrates, albeit far less processively. The single-stranded DNA binding protein RPA facilitates recombination-mediated DNA synthesis by increasing the efficiency of primer utilization, preventing polymerase stalling at specific sequence contexts, and overcoming polymerase stalling caused by topological constraint allowing the transition to a migrating D-loop. Our results support a model whereby the high-fidelity replicative DNA polymerase δ performs recombination-associated DNA synthesis, with translesion synthesis polymerases providing a supportive role as in normal replication.

Show MeSH

Related in: MedlinePlus

RPA facilitates efficient D-loop extension by Pol δ. (A) Reactions were carried out as described, with RPA titrated at the following concentrations: 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 6.0 μM. Samples (30 min time point) were analyzed by 1.2 % alkaline agarose gel electrophoresis. (B) Quantitation of synthesis products in (A). Numerical data and errors are in Supplementary Table S2. (C) Quantitation of representative sites of replication stalling (0–2 µM: average ± SEM, n = 3). (D) Reactions in (A), analyzed by 5.5 % denaturing acrylamide gel electrophoresis. Arrows indicate examples of G-tracts on the newly synthesized strand that elicit polymerase stalling that is suppressed by RPA. The initial stall region due to topological constraints is suppressed by addition of topoisomerase I (see Figure 4D) and marked on the right hand side.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3643601&req=5

gkt192-F5: RPA facilitates efficient D-loop extension by Pol δ. (A) Reactions were carried out as described, with RPA titrated at the following concentrations: 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 6.0 μM. Samples (30 min time point) were analyzed by 1.2 % alkaline agarose gel electrophoresis. (B) Quantitation of synthesis products in (A). Numerical data and errors are in Supplementary Table S2. (C) Quantitation of representative sites of replication stalling (0–2 µM: average ± SEM, n = 3). (D) Reactions in (A), analyzed by 5.5 % denaturing acrylamide gel electrophoresis. Arrows indicate examples of G-tracts on the newly synthesized strand that elicit polymerase stalling that is suppressed by RPA. The initial stall region due to topological constraints is suppressed by addition of topoisomerase I (see Figure 4D) and marked on the right hand side.

Mentions: While Pol δ is capable of limited DNA synthesis in the absence of PCNA and RFC, its efficiency has been shown to be low (39,40). Consistent with these observations and earlier studies using synthetic D-loops (14), we show that in the absence of these factors, Pol δ fails to catalyze significant DNA synthesis at D-loops (Figure 2D), and extended only 0.3% of substrate (Figure 3A and B, lane 5) with an average extension length of 5–10 nt (Figure 5D, lane 2). Single-stranded binding protein RPA is required both for recombination and efficient DNA replication (41). In the absence of RPA, D-loop formation was reduced by about one-third from ∼30 to ∼20% (Figure 2B), consistent with the previously identified role of RPA to bind to the displaced strand of the D-loop to stabilize this joint molecule (42). Omission of RPA also severely inhibited DNA synthesis by Pol δ. Pol δ extended only 7.4% of D-loops compared with 48% in the presence of RPA (Figure 3A and B, lanes 3 and 6; Figure 5D, lanes 3–10). In addition, DNA products formed were significantly shorter in the absence of RPA (Figures 2 and 3), owing to stalling induced by topological constraint and DNA sequence context, as discussed below. In the absence of RAD51, Pol δ generated a small amount of high molecular weight products that resulted from spontaneous annealing of the 93mer to the duplex DNA (Figure 2B). Pol δ contains a 3′–5′ nuclease and undergoes cycles of proofreading and DNA synthesis, a phenomenon termed idling (43), which causes loss of label in some reactions, in particular those lacking RPA, where idling is enhanced. In conclusion, we reconstituted a robust in vitro system with human proteins to analyze recombination-associated DNA synthesis from the invading strand of RAD51-mediated D-loops and show that human DNA polymerase δ efficiently extends the invading strand. The efficiency of using the invading 3′-end in the D-loop is similar to that observed with canonical primed templates (Figure 3B and Supplementary Figure S2) using the identical primer (93mer) and the same template sequence, as pBluescript and pUC19 share the same sequence for 1766 nt from the primer 3′-end. More importantly, the extension products displayed a highly similar length distribution, demonstrating that displacement synthesis is catalyzed with similar processivity as canonical primer extension by human Pol δ.


Reconstitution of recombination-associated DNA synthesis with human proteins.

Sneeden JL, Grossi SM, Tappin I, Hurwitz J, Heyer WD - Nucleic Acids Res. (2013)

RPA facilitates efficient D-loop extension by Pol δ. (A) Reactions were carried out as described, with RPA titrated at the following concentrations: 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 6.0 μM. Samples (30 min time point) were analyzed by 1.2 % alkaline agarose gel electrophoresis. (B) Quantitation of synthesis products in (A). Numerical data and errors are in Supplementary Table S2. (C) Quantitation of representative sites of replication stalling (0–2 µM: average ± SEM, n = 3). (D) Reactions in (A), analyzed by 5.5 % denaturing acrylamide gel electrophoresis. Arrows indicate examples of G-tracts on the newly synthesized strand that elicit polymerase stalling that is suppressed by RPA. The initial stall region due to topological constraints is suppressed by addition of topoisomerase I (see Figure 4D) and marked on the right hand side.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3643601&req=5

gkt192-F5: RPA facilitates efficient D-loop extension by Pol δ. (A) Reactions were carried out as described, with RPA titrated at the following concentrations: 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 6.0 μM. Samples (30 min time point) were analyzed by 1.2 % alkaline agarose gel electrophoresis. (B) Quantitation of synthesis products in (A). Numerical data and errors are in Supplementary Table S2. (C) Quantitation of representative sites of replication stalling (0–2 µM: average ± SEM, n = 3). (D) Reactions in (A), analyzed by 5.5 % denaturing acrylamide gel electrophoresis. Arrows indicate examples of G-tracts on the newly synthesized strand that elicit polymerase stalling that is suppressed by RPA. The initial stall region due to topological constraints is suppressed by addition of topoisomerase I (see Figure 4D) and marked on the right hand side.
Mentions: While Pol δ is capable of limited DNA synthesis in the absence of PCNA and RFC, its efficiency has been shown to be low (39,40). Consistent with these observations and earlier studies using synthetic D-loops (14), we show that in the absence of these factors, Pol δ fails to catalyze significant DNA synthesis at D-loops (Figure 2D), and extended only 0.3% of substrate (Figure 3A and B, lane 5) with an average extension length of 5–10 nt (Figure 5D, lane 2). Single-stranded binding protein RPA is required both for recombination and efficient DNA replication (41). In the absence of RPA, D-loop formation was reduced by about one-third from ∼30 to ∼20% (Figure 2B), consistent with the previously identified role of RPA to bind to the displaced strand of the D-loop to stabilize this joint molecule (42). Omission of RPA also severely inhibited DNA synthesis by Pol δ. Pol δ extended only 7.4% of D-loops compared with 48% in the presence of RPA (Figure 3A and B, lanes 3 and 6; Figure 5D, lanes 3–10). In addition, DNA products formed were significantly shorter in the absence of RPA (Figures 2 and 3), owing to stalling induced by topological constraint and DNA sequence context, as discussed below. In the absence of RAD51, Pol δ generated a small amount of high molecular weight products that resulted from spontaneous annealing of the 93mer to the duplex DNA (Figure 2B). Pol δ contains a 3′–5′ nuclease and undergoes cycles of proofreading and DNA synthesis, a phenomenon termed idling (43), which causes loss of label in some reactions, in particular those lacking RPA, where idling is enhanced. In conclusion, we reconstituted a robust in vitro system with human proteins to analyze recombination-associated DNA synthesis from the invading strand of RAD51-mediated D-loops and show that human DNA polymerase δ efficiently extends the invading strand. The efficiency of using the invading 3′-end in the D-loop is similar to that observed with canonical primed templates (Figure 3B and Supplementary Figure S2) using the identical primer (93mer) and the same template sequence, as pBluescript and pUC19 share the same sequence for 1766 nt from the primer 3′-end. More importantly, the extension products displayed a highly similar length distribution, demonstrating that displacement synthesis is catalyzed with similar processivity as canonical primer extension by human Pol δ.

Bottom Line: The repair of DNA breaks by homologous recombination is a high-fidelity process, necessary for the maintenance of genome integrity.Thus, DNA synthesis associated with recombinational repair must be largely error-free.Translesion synthesis polymerase eta (η) also extends these substrates, albeit far less processively.

View Article: PubMed Central - PubMed

Affiliation: Department of Microbiology & Molecular Genetics, University of California, Davis, Davis, CA 95616-8665, USA.

ABSTRACT
The repair of DNA breaks by homologous recombination is a high-fidelity process, necessary for the maintenance of genome integrity. Thus, DNA synthesis associated with recombinational repair must be largely error-free. In this report, we show that human DNA polymerase delta (δ) is capable of robust DNA synthesis at RAD51-mediated recombination intermediates dependent on the processivity clamp PCNA. Translesion synthesis polymerase eta (η) also extends these substrates, albeit far less processively. The single-stranded DNA binding protein RPA facilitates recombination-mediated DNA synthesis by increasing the efficiency of primer utilization, preventing polymerase stalling at specific sequence contexts, and overcoming polymerase stalling caused by topological constraint allowing the transition to a migrating D-loop. Our results support a model whereby the high-fidelity replicative DNA polymerase δ performs recombination-associated DNA synthesis, with translesion synthesis polymerases providing a supportive role as in normal replication.

Show MeSH
Related in: MedlinePlus