Phosphorylation of Rad55 on serines 2, 8, and 14 is required for efficient homologous recombination in the recovery of stalled replication forks - PubMed (original) (raw)

Phosphorylation of Rad55 on serines 2, 8, and 14 is required for efficient homologous recombination in the recovery of stalled replication forks

Kristina Herzberg et al. Mol Cell Biol. 2006 Nov.

Abstract

DNA damage checkpoints coordinate the cellular response to genotoxic stress and arrest the cell cycle in response to DNA damage and replication fork stalling. Homologous recombination is a ubiquitous pathway for the repair of DNA double-stranded breaks and other checkpoint-inducing lesions. Moreover, homologous recombination is involved in postreplicative tolerance of DNA damage and the recovery of DNA replication after replication fork stalling. Here, we show that the phosphorylation on serines 2, 8, and 14 (S2,8,14) of the Rad55 protein is specifically required for survival as well as for normal growth under genome-wide genotoxic stress. Rad55 is a Rad51 paralog in Saccharomyces cerevisiae and functions in the assembly of the Rad51 filament, a central intermediate in recombinational DNA repair. Phosphorylation-defective rad55-S2,8,14A mutants display a very slow traversal of S phase under DNA-damaging conditions, which is likely due to the slower recovery of stalled replication forks or the slower repair of replication-associated DNA damage. These results suggest that Rad55-S2,8,14 phosphorylation activates recombinational repair, allowing for faster recovery after genotoxic stress.

PubMed Disclaimer

Figures

FIG. 1.

Identification of phosphorylation sites on Rad55 protein. (A) Purification of Rad55 protein. Shown is a Coomassie-stained gel of 1 μg of the Rad55-Rad57 heterodimer purified from undamaged and damaged S. cerevisiae cells. −, absence of; +, presence of. (B) Schematic representation of Rad55 protein and sequence of the N terminus. The black box indicates the conserved RecA core, A and B designate the ATP binding/hydrolysis motifs (Walker boxes A and B). *, phosphorylated residues; +, the presence of phosphorylation as determined by mass spectrometry; −, the absence of phosphorylation as determined by mass spectrometry. (C) Phosphorylated peptides identified by mass spectrometry. Phosphorylated residues are in bold and underlined. (D) A drop dilution assay measured sensitivity to chronic exposure to MMS, showing complementation of MMS sensitivity of a _rad55_Δ _rad57_Δ strain (WDHY1188) by the Rad55-Rad57 overexpression plasmid in comparison to a negative control (_rad55_Δ _rad57_Δ with empty vector) and a positive control (wild-type FF18734 with empty vector) at 30°C. Plates were photographed after 3 days of incubation.

FIG. 2.

Rad55 is phosphorylated in vitro by Rad53 kinase and interacts with Rad53 in vivo. (A) Rad55 is phosphorylated in vitro by Rad53 kinase. Coomassie-stained gel (lower panel) and corresponding autoradiograph (upper panel) of in vitro kinase reactions with wild-type (lanes 1 to 3) and kinase-deficient Rad53 proteins (Rad53-kd, Rad53-K227A) (lanes 4 to 6) with no substrate (lanes 1 and 4), Rad55-Rad57 wild-type (lanes 2 and 5), or Rad55-S2,8,14A-Rad57 mutant (lanes 3 and 6) substrate. (B) In vivo interaction between Rad55 protein and Rad53 kinase. Schematic representation of Rad53 kinase with a central kinase domain and two flanking FHA domains is shown on top. Two-hybrid analysis of Rad55 was performed with wild-type and FHA domain mutant Rad53 proteins (Rad53-fha1, S85A H88A; Rad53-fha2, S619A H622A). β-gal, β-galactosidase. −, absence of; +, presence of.

FIG. 3.

rad55-S2,8,14A is sensitive to genome-wide genotoxic stress. (A) Drop dilution assay measuring sensitivity to chronic exposure to MMS of wild-type (strain WDHY2015), _rad55_Δ (WDHY2009), and rad55-S2,8,14A (WDHY2016) cells. (B) Cell survival assay measuring colony formation under MMS exposure using the strains in panel A. Given are the means and standard deviations (error bars) for three independent determinations.

FIG. 4.

rad55-S2,8,14A is proficient in repair of a single gap or DSB. (A) Repair assay for a single double-stranded DNA gap (1). The black dot indicates the relative position of the met17-s mutation. Given are the means and standard deviations of three independent determinations of wild-type (strain WDHY1800), _rad55_Δ (WDHY1999), and rad55-S2,8,14 A (WDHY1998) cells. (B) DSB repair assay with HO endonuclease. Results shown are the means and standard deviations (error bars) of three independent determinations with wild-type (WDHY2015), _rad55_Δ (WDHY2009), and rad55-S2,8,14A (WDHY2016) cells.

FIG. 5.

Rad55-S2,8,14A protein displays wild-type protein levels. Shown are the expression and protein stability of Rad55-S2,8,14A protein and immunoblot of proteins from wild-type (strain WDHY2015), Rad55-S2,8,14A (WDHY2016), and _rad55_Δ (WDHY2009) cells immunoprecipitated from cell extracts using anti-Rad55 antibodies. Cells were grown in the absence or presence of MMS (0.1%, 2 h).

FIG. 6.

Synergistic interaction of rad55-S2,8,14A with hypomorphic alleles of RAD52 and RFA1. (A) Epistasis analysis of _rad52_Δ (strain LSY718), _rad55_Δ (WDHY2009), rad52-myc18 (WDHY2060), and rad55-S2,8,14A (WDHY2016) single and double (WDHY2061 and WDHY2063) mutants for the survival of chronic exposure to MMS. (B) Epistasis analysis of _rad55_Δ (WDHY2009), rad55-S2,8,14A (WDHY2016), and rfa1-t11 (WDHY2159) single and double (WDHY2096 and WDHY2137) mutants for the survival of chronic exposure to MMS, HU, and UV. In panels A and B, the wild type (wt) was WDHY2015.

FIG. 7.

Defective recovery of stalled replication forks in rad55-S2,8,14A. (A) Scheme of replication fork arrest/recovery experiment. (B) Ethidium bromide-stained pulsed-field gels from replication fork recovery time courses with the wild type (strain WDHY2015) and rad55-S2,8,14A (WDHY2016). A, 0 min after α-factor release. Recovery time points were 0, 60, 80, 100, and 120 min, and 4 or 16 h after 1 h of MMS exposure. The arrows point to the largest chromosome (XII), which recovered starting at 80 min in the wild type but not until 4 h in rad55-S2,8,14A. (C) Quantitation of chromosome XII recovery. (D) Bulk DNA replication was monitored by flow cytometry during the course of the experiment.

FIG. 8.

rad55-S2,8,14A displays synthetic phenotypes with PRR mutants. Scheme of the postreplication repair (PRR) pathway in S. cerevisiae and epistasis analysis of _rad55_Δ (strain WDHY2009), rad55-S2,8,14A (WDHY2016), and rad18 (WDHY2018) single and double (WDHY2039 and WDHY2178) mutants for the survival of chronic exposure to MMS and HU (top panel) as well as that of _rad55_Δ, rad55-S2,8,14A, rad5 (WDHY2234), rev3 (WDHY2232), and rad30 (WDHY2236) single and double (WDHY2226, WDHY2228, WDHY2230, WDHY2238, WDHY2240, and WDHY2242) mutants for the survival of chronic exposure to MMS and HU (bottom panel). The wild type (wt) was WDHY2015.

FIG. 9.

Epistasis of rad55-S2,8,14A with DNA damage checkpoint mutants. Scheme of the DNA damage checkpoints in S. cerevisiae and epistasis analysis of _rad55_Δ (strain WDHY2213), rad55-S2,8,14A (WDHY2245), mec1 (WDHY1638), mrc1 (WDHY2277), and rad9 (WDHY2278) single and double (WDHY2255, WDHY2273, WDHY2279, WDHY2280, WDHY2285, and WDHY2287) mutants for the survival of chronic exposure to MMS (top panel) and HU (bottom panel). The wild type (wt) was WDHY1637.

FIG. 10.

Model for activation of homologous recombination at stalled replication forks and associated gaps through the phosphorylation of Rad55 protein at S2,8,14. In undamaged cells, RPA is critical for lagging strand replication and the DNA damage checkpoint is not engaged. Upon genotoxic stress by MMS, UV, or HU, replication forks stall, leading to the activation of the DNA damage checkpoint and recruitment/activation of checkpoint kinases. The Mec1 and Rad53 may recruit Rad55-Rad57 (red double star) as one of their substrates to the stalled fork and target ssDNA at this site for Rad51 filament formation (blue circles) and ensuing homologous recombination. Shown is blockage in the lagging strand (black boxes), which may allow direct resumption of replication after the lagging strand polymerase disengages from the block. It is unclear whether the fork can resume at this point or requires gap repair (or translesion synthesis) (not shown) prior to resumption. Sister gap repair by recombination converts the gap (ssDNA) to duplex DNA and the termination of checkpoint activation.

References

1. Bärtsch, S., L. E. Kang, and L. S. Symington. 2000. RAD51 is required for the repair of plasmid double-stranded DNA gaps from either plasmid or chromosomal templates. Mol. Cell. Biol. 20:1194-1205. - PMC - PubMed
1. Bashkirov, V. I., E. V. Bashkirova, E. Haghnazari, and W. D. Heyer. 2003. Direct kinase-to-kinase signaling mediated by the FHA phosphoprotein recognition domain of the Dun1 DNA damage checkpoint kinase. Mol. Cell. Biol. 23:1441-1452. - PMC - PubMed
1. Bashkirov, V. I., K. Herzberg, E. Haghnazari, A. S. Vlasenko, and W. D. Heyer. 2006. DNA-damage induced phosphorylation of Rad55 protein as a sentinel for DNA damage checkpoint activation in S. cerevisiae. Methods Enzymol. 409:166-182. - PubMed
1. Bashkirov, V. I., J. S. King, E. V. Bashkirova, J. Schmuckli-Maurer, and W. D. Heyer. 2000. DNA repair protein Rad55 is a terminal substrate of the DNA damage checkpoints. Mol. Cell. Biol. 20:4393-4404. - PMC - PubMed
1. Bianco, P. R., R. B. Tracy, and S. C. Kowalczykowski. 1998. DNA strand exchange proteins: a biochemical and physical comparison. Front. Biosci. 3:570-603. - PubMed

Phosphorylation of Rad55 on serines 2, 8, and 14 is required for efficient homologous recombination in the recovery of stalled replication forks - PubMed (original) (raw)