The Yeast DNA Damage Checkpoint Kinase Rad53 Targets the Exoribonuclease, Xrn1

Abstract

The highly conserved DNA damage response (DDR) pathway monitors the genomic integrity of the cell and protects against genotoxic stresses. The apical kinases, Mec1 and Tel1 (ATR and ATM in human, respectively), initiate the DNA damage signaling cascade through the effector kinases, Rad53 and Chk1, to regulate a variety of cellular processes including cell cycle progression, DNA damage repair, chromatin remodeling, and transcription. The DDR also regulates other cellular pathways, but direct substrates and mechanisms are still lacking. Using a mass spectrometry-based phosphoproteomic screen in Saccharomyces cerevisiae, we identified novel targets of Rad53, many of which are proteins that are involved in RNA metabolism. Of the 33 novel substrates identified, we verified that 12 are directly phosphorylated by Rad53 in vitro: Xrn1, Gcd11, Rps7b, Ded1, Cho2, Pus1, Hst1, Srv2, Set3, Snu23, Alb1, and Scp160. We further characterized Xrn1, a highly conserved 5' exoribonuclease that functions in RNA degradation and the most enriched in our phosphoproteomics screen. Phosphorylation of Xrn1 by Rad53 does not appear to affect Xrn1's intrinsic nuclease activity in vitro, but may affect its activity or specificity in vivo.

Keywords: DNA Damage Response; Rad53; Xrn1; checkpoint; phosphoproteomics.

Figure 1

Phosphoproteomic screen for Rad53 targets. A. Western blots showing mobility shift of Rad53 substrates: Sld3, Ndd1, and Dbf4 in the strain indicated. ppp∆ refers to ptc3∆ ptc2∆ pph3∆. Sld3 and Ndd1 are respectively tagged with MYC and Flag epitope tags and visualized with antibodies against the tag. Dbf4 is visualized with antibodies against Dbf4. B. Schematic of the experiment for phosphopeptide mass spectrometry. Both strains are in the sml1∆ ptc3∆ ptc2∆ pph3∆ background. Asynchronous yeast cultures were treated with galactose for 1 hr to induce Rad53 expression followed by the addition of 2 μg/mL 4-NQO for 15 min before harvesting. C. Comparison of fold changes for a phosphopeptide between replicates. Box region shows phosphopeptides that are ≥25 fold enriched in both replicates in GAL1pr-RAD53 TEF1pr-DUN1 strains. D. Enriched biological process-associated GO terms using DAVID for the top 54 proteins.

Figure 2

Schematic of RNA processing pathways. A. Schematic of mRNA dynamics in the cell. mRNA is dynamically regulated in the cell, and can be translated into proteins, stored in stress granules, degraded in the processing bodies (P-bodies), or undergo mRNA decay outside of the P-bodies. Proteins whose phosphopeptides were ≥25 fold enriched in our phosphoproteomics were placed in the processes in which they are known to be involved (in blue). *Dcp2 is 20X enriched in GAL1pr-RAD53 vs. rad53∆. B. Schematic of rRNA processing in the cell. The 35S pre-rRNA transcript undergo a series of distinct processing events to generate the rRNA components of the 40S ribosome (18S) and 60S ribosome (5.8S and 25S). The 5S component of the 60S ribosome is independently transcribed and not shown. Proteins whose phosphopeptides were ≥25 fold enriched in our phosphoproteomics are indicated (in blue).

Figure 3

Downregulation of rRNA processing or protein synthesis is independent of the DNA damage checkpoint. A. Wild type, mec1∆, and rad53∆ cells are untreated or treated with 2 μg/mL 4-NQO. Samples were collected at 30 min and 60 min after treatment for RNA extraction. The full length 35S pre-rRNA transcript is processed to yield mature 18S, 5.8S, and 25S rRNA. Northern blot with probe against full-length 5.8S sequence allows detection of the 32/33S and 27S intermediates, and the fully processed 5.8S rRNA. B. Film of S³⁵-methionine incorporation. Wild type, rad53∆, and mec1∆ tel1∆ cells were untreated or treated with 2 μg/ml 4-NQO for 15 min S³⁵-methionine was added and samples were taken after 1 min and 5 min. C. Film of S³⁵-methionine incorporation. Left panel: wild type and gcn2∆ cells were untreated or treated with 2 μg/ml 4-NQO for 60 min Middle panel: G1 arrested wild type and gcn2∆ cells were untreated or treated with 0.05% MMS for 60 min Right panel: G1 arrested wild type and cdc13-1 cells were grown at 23°C or 32°C for 2 hr, the permissive and nonpermissive temperature for cdc13-1 respectively. S³⁵-methionine was added and samples were taken after 1 min and 5 min. D. Western blot of eIF2 alpha phosphorylation. Left planel: wild type and gcn2∆ cells were untreated or treated with 2 μg/ml 4-NQO for 60 min Right panel: G1 arrested wild type and cdc13-1 cells were grown at 23°C or 32°C for 2 hr. Samples were blotted for phosphorylated eIF2 alpha-S51 and Dbf4. E. Wild type and rad53∆ cells were arrested in G1 with α factor and released into rich media or rich media with 200 mM HU for 45 min before harvesting. Total RNA and ribosome footprints were purified using Illumina’s ARTseq kit before sequencing. Left panel is an M-A plot showing the average expression for a gene between rad53∆ and wild type cells (x-axis) vs. the fold change between rad53∆ and wild type cells (y-axis) in the absence of DNA damage. Right panel is an M-A plot showing the average expression for a gene between rad53∆ and wild type cells (x-axis) vs. the fold change between rad53∆ and wild type cells (y-axis) in the presence of HU.

Figure 4

Validation by in vitro kinase assay of proteins associated with enriched phosphopeptides in the screen. A. In vitro kinase assay for 44 proteins whose phosphopeptides were enriched by ≥25 fold and where the GFP-tagged strains were available in the GFP collection. For each protein set, - indicates no kinase control, R indicates Rad53, and D indicates Dun1. Top panels are autorads showing ³²P signal indicating transfer of ³²P-ATP to substrate. * indicates Rad53 autophosphorylation signal. ^† indicates contaminating kinase in IP. Bottom panels show Western blot for GFP as loading control for each lane. ^‡ Dcp2 (20X enriched) and Pbp1 (2X enriched) were selected because of their connection to Xrn1 in P body biology.

Figure 5

Rad53 is required for Xrn1 mobility shift during DNA damage. A. Western blot showing Xrn1-GFP shift upon DNA damage, and is dependent on Rad53 and Mec1/Tel1. Asynchronous yeast cultures were treated with 2 μg/mL 4-NQO for 90 min before harvesting. B. Western blot showing Xrn1-GFP shift is independent of Dun1. Asynchronous cells were treated with 0.05% MMS for 3 hr before harvesting. C. Western blot showing Xrn1-Flag shift is dependent on Rad53, but independent of Dun1. Asynchronous cells were treated with 2 μg/mL 4-NQO for 90 min. Rad53 is shown as a control for DNA damage treatment, because it hyper-shifts in response to DNA damage.

Figure 6

Xrn1 phosphorylation does not interfere with its nuclease activity in vitro, or its ability to degrade endogenous transcripts. A. Nuclease assay using in vitro transcribed RNA as substrate with purified Xrn1 from cells untreated (-) and treated (+) with 2 μg/mL 4-NQO for 1 hr. Nuclease assay was allowed to proceed for 60 min and samples were collected at the indicated time. Mock IP sample was allowed to proceed for 60 min B. Quantification of product (NMP) formation normalized to t = 0 signal. C. Western blot showing relative Xrn1 abundance used in experiment in A. D. Normalized mRNA levels of GAL1 transcript turnover. The GAL1 promoter is rapidly repressed when galactose in the media is replaced with glucose. 4-NQO was added to a concentration of 2 μg/mL concurrently with glucose and samples were collected at the indicated time. E. Normalized mRNA levels of HST3 transcript turnover. HST3 is placed under the GAL1 promoter, which is downregulated in glucose and the rate of degradation of HST3 was determined. 4-NQO was added to a concentration of 2 μg/mL concurrently with glucose and samples were collected at the indicated time. F. Tenfold serial dilutions of the indicated strains grown on YPD, YPD + 100 mM HU, or YPD + 1 μg/mL phleomycin (or YPD + 1.5 μg/mL phleomycin). Plates were grown at 30°C for 2-3 days before scanning. G. Tenfold serial dilutions of the indicated strains grown on YPD, YPD + 100 mM HU, or YPD + 1 μg/mL phleomycin. Plates were grown at 30°C for 2-3 days before scanning.

Figure 7

Change in Xrn1 interactions in rad53∆ cells. A. Co-IP of Xrn1-Flag and Lsm3-GFP in wild type or rad53∆ cells. Asynchronous cells were untreated (-) or treated (+) with 0.05% MMS for 3 hr. Samples were immunoprecipitated with Flag antibodies. B. List of proteins identified in reciprocal SILAC mass spectrometry analyses of Xrn1-Flag IP from wild type and rad53∆ cells treated with 2 μg/mL 4-NQO for 60 min. C. Co-IP of Xrn1-Flag and Mss116-GFP or YMR196W-GFP in wild type or rad53∆ cells, as indicated. Asynchronous cells were untreated (-) or treated (+) with 0.05% MMS for 3 hr, and samples were immunoprecipitated with Flag antibodies.