Cyclin-dependent kinase modulates budding yeast Rad5 stability during cell cycle

Abstract

The DNA damage tolerance (DDT) pathway facilitates the bypass of the fork-blocking lesions without removing them through either translesion DNA synthesis or error-free damage bypass mechanism. The Saccharomyces cerevisiae Rad5 is a multi-functional protein involved in the error-free branch of the DDT pathway, and its protein level periodically fluctuates through the cell cycle; however, the mechanistic basis and functional importance of the Rad5 level for the cell cycle regulation remain unclear. Here, we show that Rad5 is predominantly phosphorylated on serine 130 (S130) during S/G2 phase and that this modification depends on the cyclin-dependent kinase Cdc28/CDK1. We also show that the phosphorylated Rad5 species at S130 exhibit a relatively short half-life compared with non-phosphorylated Rad5 moiety, and that the Rad5 protein is partially stabilized in phosphorylation-defective rad5 S130A cells. Importantly, the elimination of this modification results in a defective cell-cycle dependent Rad5 oscillation pattern. Together, our results demonstrate that CDK1 modulates Rad5 stability by phosphorylation during the cell cycle, suggesting a crosstalk between the phosphorylation and degradation of Rad5.

Fig 1. Rad5 phosphorylation under normal growth conditions.

(A) MMS sensitivity of wild-type, rad5Δ and RAD5-Myc cells. Ten-fold serial dilutions of asynchronous cell cultures were spotted onto YPDA plates with or without MMS (0.01%) and incubated at 30°C for 3 days. (B) Detection of Rad5 phosphorylation by western blotting with anti-c-Myc antibody. Log-phase cultures of wild-type RAD5-Myc cells were either mock treated (-) or treated with 0.1% MMS (+) for 1 h. Protein extracts were then processed on a standard SDS-PAGE gel and on a Phos-tag SDS-PAGE gel. Rad5 and Rad53 proteins were detected with anti-c-Myc and anti-Rad53 antibodies, respectively. The slower migrating bands are indicated by arrow heads. (C) Immunoprecipitated Rad5 was treated with lambda phosphatase (or mock treatment) and phosphatase inhibitors prior to Phos-tag SDS-PAGE, followed by immunoblotting with anti c-Myc antibody.

Fig 2. Identification of phosphorylation sites in Rad5.

(A) Schematic representation of the Rad5 domains, including the HIRAN (176–285), helicase (440–848 and 1030–1164) and RING finger (913–961) domains, which are shown in the left panel. N- and C-terminal truncations of Rad5 shown in the left panel were constructed by fusion to a Myc epitope tag. rad5Δ cells were transformed with each of the CEN/ARS plasmids carrying wild-type or truncated alleles of RAD5. Expression of wild-type and truncated alleles from the endogenous RAD5 promoter was confirmed by Phos-tag western blotting using c-Myc antibody (right panel). (B) N-terminal truncations ranging in size from 24 to 174 residues, shown in the left panel, were constructed by fusion to a Myc epitope tag. Expression of truncations was examined as described in (A) (right panel). (C) Phosphorylation state of Rad5-Myc with each single or double amino acid substitutions in the N-terminal region. A map of the Rad5 N-terminal region between amino acids 114 and 144 is shown in the left panel. Asterisks indicate the putative phosphorylation sites, S129 and S130. rad5Δ cells were transformed with plasmids carrying each rad5 mutant. Conserved amino acid residues among budding yeast are shown in bold. Phosphorylation patterns were analyzed as described in (A) (right panel). Tubulin served as a loading control.

Fig 3. Functional analysis of Rad5 phosphorylation in DNA damage tolerance pathways.

(A) DNA damage sensitivity of rad5 mutants. rad5Δ cells carrying each of the indicated plasmids were 10-fold serially diluted and exposed to the indicated DNA-damaging agents, and plates were incubated at 30°C for 3 days. (B) Mutation frequencies at the CAN1 locus in rad5Δ cells carrying the indicated plasmids. Cells were grown in SC-LEU medium at 30°C. Canavanine-resistant mutants were selected on synthetic complete medium lacking leucine and arginine, and containing canavanine. Error bars indicate the standard errors of the three independent experiments.

Fig 4. Elevated phosphorylation levels of Rad5 at S130 during the S to G2/M phases.

(A, B) RAD5-Myc cells grown to early log phase at 30°C were arrested in G1 with α-factor for 2 h. Cells were then released into fresh YPDA medium. Samples from synchronized cell cultures were taken at the indicated time points after release from G1 block. Cell-cycle progression was determined by flow cytometry (A) and Rad5 protein was analyzed by Phos-tag western blotting (B). Tubulin served as a loading control. Asn denotes asynchronously growing cells. (C) Total amount of Rad5 is shown as relative band intensities together with the ratio of the relative amount of the two bands. Values are expressed relative to 1.00 for total amount of Rad5 at time zero. Error bars are derived from standard errors of the three independent experiments. (D, E) RAD5-Myc, rad5-S130A and rad5-S129A S130A cells were grown to log phase, arrested at G1 by α-factor, and then released into fresh YPDA medium. Cell cycle progression was analyzed by flow cytometry (D) and Rad5 protein levels were analyzed by Phos-tag western blotting (E). (F) cdc28-as1 RAD5-Myc cells were grown to log phase or arrested at S phase with 200 mM HU or at G2 phase with 20 μg/ml nocodazole. After 2 h, cultures (pre-treatment) were divided equally and treated with DMSO (mock) or 5 μM 1NM-PP1 for 1 h. Rad5 protein was analyzed by Phos-tag western blotting.

Fig 5. Rapid decay of S130 phosphorylation species.

(A) RAD5-Myc cells were asynchronously grown in YPDA, and were then treated with CHX (400 μg/ml) for the indicated times to shut off transcription. The stability of Rad5 was examined by Phos-tag western blotting. (B) The band intensities of phosphorylated and unphosphorylated Rad5 shown in (A) were quantified (n = 3). Normalized levels of each band relative to 1.0 at time 0 are shown. Error bars represent the standard errors of the three independent experiments. (C) Cells grown to early log phase at 30°C were transferred into YPDA containing CHX. Samples were obtained at 0, 30, and 90 min after the addition of CHX, and were then subjected to Phos-tag western blotting. (D) The overall band intensity of each lane in (C) was quantified, and the Rad5 protein level remained after the addition of CHX was shown relative to 1.0 at time 0. Error bars represent the standard errors of the three independent experiments.