autophagy-dependent regulation of the dna damage …dna damage and cellular stress responses...

DNA Damage and Cellular Stress Responses

Autophagy-Dependent Regulation of the DNA DamageResponse Protein Ribonucleotide Reductase 1

Madhu Dyavaiah1,2, John P. Rooney1,2, Sridar V. Chittur1,2, Qishan Lin1,2, and Thomas J. Begley1,2

AbstractProtein synthesis and degradation are posttranscriptional pathways used by cells to regulate protein levels. We

have developed a systems biology approach to identify targets of posttranscriptional regulation and we haveemployed this system in Saccharomyces cerevisiae to study the DNA damage response. We present evidence that50% to 75% of the transcripts induced by alkylation damage are regulated posttranscriptionally. Significantly, wedemonstrate that two transcriptionally-induced DNA damage response genes, RNR1 and RNR4, fail to showsoluble protein level increases after DNA damage. To determine one of the associated mechanisms ofposttranscriptional regulation, we tracked ribonucleotide reductase 1 (Rnr1) protein levels during the DNAdamage response. We show that RNR1 is actively translated after damage and that a large fraction of thecorresponding Rnr1 protein is packaged into a membrane-bound structure and transported to the vacuole fordegradation, with these last two steps dependent on autophagy proteins. We found that inhibition of target ofrapamycin (TOR) signaling and subsequent induction of autophagy promoted an increase in targeting of Rnr1 tothe vacuole and a decrease in soluble Rnr1 protein levels. In addition, we demonstrate that defects in autophagyresult in an increase in soluble Rnr1 protein levels and a DNA damage phenotype. Our results highlight roles forautophagy and TOR signaling in regulating a specific protein and demonstrate the importance of these pathwaysin optimizing the DNA damage response. Mol Cancer Res; 9(4); 462–75. �2011 AACR.

Introduction

Cellular exposure to genotoxic agents can result in DNAdamage and promote mutations and cell death. In responseto DNA damage, cells activate genome maintenance path-ways associated with the DNA damage response. In Sac-charomyces cerevisiae, the DNA damage response is activatedby Mec1, Rad53, and Dun1 kinases to mediate cell-cyclecheckpoint activation, to activate DNA repair and topromote the regulation of ribonucleotide reductase(RNR) activity (1, 2). RNR catalyzes the rate-limiting stepin the production of dNTPs. RNR activity is high during S-phase and increases after DNA damage to elevate dNTPlevels (3, 4). In S. cerevisiae, there are 4 RNR genes encoding2 large subunits (RNR1 and RNR3) and 2 small subunits(RNR2 and RNR4), with the final RNR complex consistingof a tetramer of 2 large and 2 small subunits. To promote

high RNR activity after DNA damage, the transcription ofthe RNR subunits RNR1, RNR3, and RNR4 is induced in aMec1-dependent fashion (5). Similarly, after DNA damage,RNR activity is increased by Dun1-dependent phosphor-ylation of the Rnr1 inhibitor Sml1 (6). RNR activity hasproved to be a vital regulator of DNA synthesis and cell-cycle progression after DNA damage; RNR regulation bytranscription and protein kinase cascades can be consideredprototypical for components of the DNA damage response.After DNA damage, transcription and kinase-based sig-

naling are global themes: the DNA damage response worksin tandem with these and a broad range of other cellularprocesses to optimize DNA repair. Transcriptional profilingstudies of S. cerevisiae have demonstrated that alkylationdamage induces the transcription of hundreds of genescorresponding to proteins associated with DNA repair,cell-cycle checkpoints, amino acid metabolism and proteindegradation (7–11). Protein degradation has been observedpreviously for components of the DNA damage response(12, 13) and, in theory, can work to optimize essentialenzyme activity. In addition, protein degradation is used toremove damaged or unfolded proteins. Protein degradationcan occur using either ubiquitin-proteasome or vacuolar-mediated pathways. Both proteasome and vacuolar-mediated protein degradation have been implicated in stressresponses and components of both pathways are associatedwith the reaction to and recovery from DNA-damagingagents (14–16). Transcriptional profiling and targeted stu-dies have highlighted the role of the proteasome after

Authors' Affiliations: 1Department of Biomedical Sciences; and 2CancerResearch Center, University at Albany, State University of New York,Rensselaer, New York

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

Corresponding Author: Thomas J. Begley, University at Albany, StateUniversity of New York, Rensselaer, New York 12144. Phone: 518-591-7153; Fax: 518-525-2799; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-10-0473

�2011 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 9(4) April 2011462

cellular exposure to alkylating agents and demonstrated thatthe proteasome-associated factor Rpn4 is involved in theregulation of the DNA damage response (9). Global phe-notypic and subcellular localization studies have shown thatthe vacuolar Hþ–ATPase complex modulates the toxicity ofthe alkylating agent MMS (17, 18). Phenotypic studies alsodemonstrated that the vacuole plays an important role inpromoting cellular viability after S. cerevisiae have beenexposed to bleomycin (19).The ability of the vacuole to modulate the toxicity of

damaging agents may rest either in its role in the sequestra-tion of toxic compounds or in its degradation activity. Thedegradation of cellular constituents in the vacuole requires afunctional set of vacuolar ATPases, degradation enzymesand autophagy proteins. Yeast mutants that are compro-mised for vacuole function are defective in autophagy anddisplay reduced degradation of proteins and organelles (20).Autophagy is a complex catabolic program that uses thevacuole and lysosome in yeast and humans, respectively. Inthe process of autophagy, parts of the cytoplasm, long-livedproteins, and intracellular organelles are sequestered into amembrane-bound structure termed the autophagosome.The autophagosome is then targeted to the vacuole fordegradation (21). TOR, a phosphatidylinositol kinase regu-lated by nutritional stress, controls autophagy. TOR signal-ing is active under nutrient-rich conditions and is dormantunder starvation conditions to induce autophagy, therebypromoting the breakdown and recycling of cellular macro-molecules (22, 23). In addition to controlling autophagy,TOR regulates cell growth, cell cycle, protein synthesis, andnutrient import. In mammals, autophagy is regulated bymTOR and this catabolic process has been implicated incarcinogenesis and cancer progression (24, 25). The humanBeclin 1 gene, an ortholog of yeast ATG6, is a criticalcomponent of the mammalian autophagy pathway and isclassified as a haploinsufficient tumor suppressor (26).Beclin 1 is monoallelically deleted in breast, ovarian, andprostate cancers, and beclin-1 knockout mice showincreased levels of spontaneous tumors (27, 28). Allelic lossof Beclin 1 leads to genome instability (29), but themechanism connecting autophagy to genome maintenanceis not well understood.In this study, we postulated that many DNA damage

response transcripts are regulated posttranscriptionally. Totest our hypothesis, we have used transcriptional profilingand proteomic approaches to identify targets of posttran-scriptional regulation following DNA damage. We deter-mined that over 50% of the transcripts induced by MMSwere regulated after transcription, with RNR1 and RNR4as specific examples. We have used genetic and pharma-cological methods to demonstrate increased RNR1 tran-scription, increased Rnr1 protein synthesis, increasedRnr1 targeting to a membrane-bound structure andincreased Rnr1 degradation after DNA damage. Notably,we also demonstrate that Rnr1 targeting and degradationincreased upon TOR inhibition, and in all cases theremoval of Rnr1 from the soluble fraction required pro-teins participating in autophagy. We show that cells

defective in autophagy have increased Rnr1 protein levelsand they display a DNA damage phenotype under con-ditions of TOR inhibition. Under nutrient-limiting con-ditions we propose that the autophagy-dependentdegradation of Rnr1 serves to optimize the DNA damageresponse. Ultimately, the following results highlight newregulatory roles for autophagy and TOR signaling duringthe DNA damage response and further links nutrientsensing to the control of RNR activity.

Materials and Methods

Yeast and growth conditionsSupplementary Table 5S lists the strains and oligonucleo-

tides used in this study. All mutants were made using theircorresponding G418 knockout cassette from the Sacchar-omyces Gene Deletion Project and were selected on YPDplates supplemented with G418 (200 mg/mL). Mutantswere confirmed by PCR. Media preparation and other yeastmanipulations were performed using standard methods.The nutritional stress medium consisted of 0.17% yeastnitrogen base without amino acids, ammonium sulfate and2% glucose.

RNA analysisTotal RNA was isolated using the RNeasy Mini Kit

(Qiagen) with 250 U of Zymolyase (Associates of CapeCod). RNA was isolated from the spheroplasts and exam-ined spectroscopically, by agarose gel electrophoresis andBioanalyzer analysis (Agilent). Northern blot analysis wasperformed using 10 to 12 mg of RNA with detectionfacilitated using the Chemiluminescent Nucleic AcidDetection Module (Pierce). Affymetrix Gene Chip (Yeast2.0) analysis was performed as previously described (30).RNA from untreated and MMS-treated cells (n ¼ 2) waslabeled, applied to chips, and the resulting data analyzedusing CyberT software (31). Polysome profiles were per-formed as previously described (32).

Protein analysisPelleted cells were lysed with 50 mL of a boiling SDS

solution (50 mmol/L Tris-HCl, pH 7.5, 5% SDS, 5%glycerol, 50 mmol/L DTT, 5 mmol/L EDTA, 2 mg/mLLeupeptin, 2 mg/mL Pepstatin A, 1 mg/mL Chymostatin,0.15 mg/mL Benzamidine, 0.1 mg/mL Pefabloc, 8.8 mg/mL Aprotanin, and 3 mg/mL Anitpain). Lysed cells werecentrifuged for 5 minutes at 13,000 rpm and the resultingsupernatant was used as protein extract. Membrane proteinwas extracted using Mem-PER Eukaryotic Membrane Pro-tein Extraction Kit (Pierce). Protein concentrations weremeasured by NanoDrop 1000 (Thermo Scientific) usingBSA as a standard. Western blots were performed asdescribed (33) with an anti-TAP antibody (Open Biosys-tems), anti-Rnr1 antibody (Santa Cruz Biotechnology) andanti-b-tubulin antibody (Abcam). Individual bands werequantitated using AlphaEaseFC software (Imgen Technol-ogies). iTRAQ analysis was performed using a methodologypreviously reported (34, 35).

Rnr1 Regulation by Autophagy

www.aacrjournals.org Mol Cancer Res; 9(4) April 2011 463

Protein degradation analysisRnr1 protein degradation was measured in the presence

of either the proteasome inhibitor MG132 (Sigma) or thevacuolar inhibitor PMSF (Pierce) in erg6D cells. Cultureswere grown at 30�C to 5� 106 cells/mL in YPDmedia andwere preincubated with either 100 mmol/L MG132 or1 mmol/L PMSF as described (14).

High-throughput ECL analysis of TAP-tagged proteinsLysates from MMS-treated and untreated cells were added

to the High Bind Plate from Meso Scale Discovery (MSD).Plates were incubated at room temperature for 1 hour,blocked, and washed according to manufacturer instructions.A 1:500 dilution of Anti-tap antibody was added followedby an MSD Sulfo-tag labeled secondary antibody. The platewas washed with TBST and read immediately after addingMSD Read Buffer using the SECTOR 2400 from MSD.

Fluorescence and confocal microscopy and flowcytometryFluorescence microscopy was performed using an endo-

genously expressed Rnr1-GFP fusion protein (Invitrogen).Cells were grown to mid-log phase in YPDmedium, MMS-treated as indicated, and examined on a Nikon EclipseTS100 fluorescence microscope. Images were taken with aRT3 camera (Diagnostic Instruments) and SPOT Basic andImageJ software were used to capture images and quantitatethe focus area, respectively. Statistical significance wasdetermined using a t test. The vacuolar membrane wasstained using FM4-64 (Invitrogen) following the manufac-turers protocol. For confocal microscopy, the vacuolarlumen was stained using CellTracker Blue (Invitrogen)and observed under a Leica Confocal microscope usingGFP and DAPI channels. The DNA content of untreated,alpha factor synchronized, and MMS-treated cells weremeasured by flow cytometry after PI staining, as described(36).

Results

Matched systems studies identify targets forposttranscriptional regulationPrevious studies in S. cerevisiae have reported that

MMS induces hundreds of different transcripts, depend-ing on the exposure conditions (7–11). We hypothesizedthat many of the MMS-induced transcripts would beposttranscriptionally regulated, and we have matchedtranscriptional profiling and protein level data to identifypotential targets. Mild exposure to the DNA damagingagent MMS promotes a delay in cell-cycle progression,induces minimal cell death, and promotes global repro-gramming of the yeast transcriptome (37–39). We haveperformed transcriptional profiling studies to identifytranscripts regulated after MMS exposure. Similar toresults reported in the literature, Affymetrix mRNAanalysis indicated that MMS exposure induced the levelsof 145 transcripts (2.0-fold, P < 0.05) and decreased thelevels of 44 transcripts (2.0-fold, P < 0.05) after DNA

damage (Fig. 1A, Supplementary Table S1). Specifically,MMS exposure increased the levels of the DNA damageresponse transcripts RNR1, RNR3, RNR4, and MAG1(Fig. 1B). Similar to previous studies (7–11), we deter-mined that the 145 MMS-induced transcripts corre-sponded to protein activities overrepresented (P <0.05) in the following functional categories: DNAdamage response, deoxyribonucleotide metabolism,DNA repair, cell-cycle arrest, and protein/peptide degra-dation (Supplementary Table S2).To determine whether an increase in a specific tran-

script was matched at the protein level, we used globalproteomic and targeted protein analyses. We matchedprotein expression changes to our 145 MMS-inducedtranscripts. A global proteomic analysis of soluble proteinswas performed with proteins isolated from both MMS-treated and untreated cells using tandem mass spectro-metry of trypsinized total proteins. The correspondingpeptides were labeled with isobaric tags for relative andabsolute quantitation (iTRAQ) reagents. iTRAQ identi-fied peptides corresponded to approximately 600 proteins(�10% coverage), with 33 proteins upregulated (2.0-fold)and 13 proteins downregulated (2.0-fold) in the MMS-treated samples (Supplementary Table S3). Upregulatedproteins were overrepresented in the functional categoriesof stress response, translation initiation, and oxidativestress (P < 0.05, Supplementary Table S4). We pairedthe 145 damage-induced transcripts with total proteinsidentified by iTRAQ and identified 16 transcripts withcorresponding protein data (Supplementary Table S3).In this limited overlap, only 25% of the MMS-inducedtranscripts displayed concordant protein level increases,with the other 75% displaying no change or a decrease atthe protein level. The identification of protein levelchanges corresponding to the 129 other MMS-inducedtranscripts was not feasible at this point as matchingpeptides were not identified by mass spectrometry. Tohelp fill in the missing information, we used a high-throughput ECL-based technology from MesoScale Dis-coveries and analyzed 29 specific TAP-tagged strains thatcorresponded to the transcripts with the highest MMSinduction (Fig. 1A). These strains are specific to individualTAP-tagged proteins and allow for the measurement ofendogenous protein levels, using a single antibody. Wealso analyzed Rnr4-TAP as it was transcriptionallyinduced by MMS and two other components of theRNR complex were picked for analysis using the ECL-based method. Our targeted analysis of the 30 TAP-taggedproteins found only 50% of the MMS-induced transcriptsto be concordant with protein level increases. This findingfurther demonstrates that a large number of MMS-induced transcripts or corresponding proteins were regu-lated posttranscriptionally, translationally or by proteindegradation. We note that a schematic detailing thesetranscript and protein based methodologies and results isprovided in Supplementary Figure S1A.In evaluating the DNA damage response, we have deter-

mined that 2 transcriptionally-induced components fail to

Dyavaiah et al.

Mol Cancer Res; 9(4) April 2011 Molecular Cancer Research464

show a corresponding protein level increase after DNAdamage: the large and small RNR subunits Rnr1 andRnr4 (Fig. 1B). We also analyzed the MMS-inducedtranscripts ARG3 and PAC11, and corresponding proteins(Supplementary Fig. S1B), and determined that both failedto have concurrent protein level increases. This suggests thatthese transcripts were regulated posttranscriptionally. Asboth Rnr1 and Rnr4 are recognized components of theDNA damage response, we were intrigued by the discre-pancy at the protein level. Thus we keyed in on componentsof the RNR complex for the rest of our study and to probethe posttranscriptional regulation of DDR components. We

note that our assay only detected levels of soluble Rnr1 andRnr4 proteins. In addition, we only detected a slightincrease in Rad53 protein levels after MMS damage, some-what matching the transcriptional increase. However, ourprotein level analysis may have been distorted by extensivephosphorylation of Rad53 after MMS damage, which weobserved as a high-molecular-weight species on Westernblots (Supplementary Fig. S1C). Thus, we have omittedRad53 from our list of proteins decoupled from a transcrip-tional increase. Northern and Western blot analysis(Fig. 1B) of RNR1 and RNR4 transcripts and correspondingproteins further demonstrated that the MMS-induced

Figure 1. Some protein levelchanges are decoupled fromMMS-induced transcriptionalincreases. S. cerevisiae (BY4741)cultures were grown at 30� C to5 � 106 cells/mL in YPD and wereeither left untreated or exposed to0.0125% MMS, or otherconcentrations where indicated,for 1 hour. A, Upregulatedtranscripts (>2.0-fold, P < 0.05)were matched to protein levelchanges as measured by ECLanalysis of TAP-tagged proteins.B, Northern blot analysis of 4 DNAdamage-induced transcripts andWestern blots to correspondingTAP- tagged proteins wereperformed as described inMaterials and Methods. C, cellviability analysis of wild-typeBY4741 cells after 1-hourtreatment with MMS. D, Northernand Western blot analysis ofRNR1 and Rnr1 1 hour aftertreatment with MMS. E, Northernand Western blot analysis ofRNR1 and Rnr1-TAP before (Unt.)and after MMS or bleomycin(BLM) treatment were performedas described previously.

3

4

5Transcript

Protein

A

BC

0

1

2

3

-2

-1

RN

R3

FP

R1

LA

P4

HS

P12

EC

M4

GP

M2

HS

P26

GP

G1

GR

E2

CO

S3

AR

G3

GA

D1

ICY

1C

DC

91T

HI2

RN

R1

HS

P31

GS

P2

HX

K1

YB

L11

1CR

AD

51D

CS

25P

EP

3A

MS

1R

AD

53P

AC

11R

PL

30

HU

G1

AR

O1

Lo

g2

ACT1

RNR1

MMS - + - +

TubulinRnr1-TAP

1101009080706050

ACT1

RNR4

RNR3

ACT1

Rnr3-TAP

Rnr4-TAP

Tubulin

MMS (%)

D MMS (%)

40

3020100

0 0.006 0.0125 0.0025

% C

ell s

urvi

val

Transcripts Proteins

ACT1

MAG1

ACT1

Tubulin

Tubulin

Mag1-TAP

Rnr1

Tubulin

Unt. 0.006 0.0125 0.025

RNR1

ACT1 Tra

nsc

rip

tsP

rote

ins

Unt. MMS BLM

RNR1

ACT1

Rnr1

Tubulin

Unt. MMS BLM

E

ACT1 Tubulin

Transcripts Proteins

fold

ch

ang

e (M

MS

/Un

trea

ted

)



transcriptional increases for each do not lead to an increasein soluble protein. In contrast, we detected matched tran-script and protein level increases for the DNA damageresponse genes RNR3 and MAG1 (Fig. 1B). These resultssuggest that posttranscriptional regulation of the RNR1 andRNR4 transcripts or corresponding proteins is occurring.To further analyze the potential posttranscriptional reg-

ulation of RNR components after DNA damage, we haveperformed detailed analysis of RNR1 transcript and proteinlevels over 3 doses of MMS (0.006%, 0.0125%, and0.025% MMS). Cell viability results demonstrated thatthere was little killing using these MMS doses (Fig. 1C).Notably, we observed transcriptional induction of theRNR1 transcript using all 3 concentrations of MMS, yetthere was no observable increase in Rnr1 protein levels inany of the exposed cells (Fig. 1D). Thus our observationthat the transcriptional increase in RNR1 is decoupled fromprotein levels increases holds true for both sublethal andmildly lethal MMS exposures. MMS is a classic DNAdamaging agent that promotes strand breaks and replicationblocks, as well as readily methylating RNA and proteins. Torule out MMS-induced protein damage as a cause for ourobserved discrepancy, we have also analyzed the levels of theRNR1 transcript and Rnr1 protein following bleomycin-induced double-strand breaks. Similar to our MMS studies,we found that RNR1 transcript levels increased in responseto bleomycin treatment; yet, Rnr1 protein levels remainedunchanged (Fig. 1E). The transcriptional increase in RNR1after bleomycin treatment is more pronounced comparedwith MMS treatment, suggesting that the decoupling ofprotein level increases from transcriptional increases is moredramatic after treatment with agents that directly causedouble strand breaks. The MMS and bleomycin resultssupport our hypothesis that protein-level changes can bedecoupled from transcriptional changes during the DNAdamage response and that posttranscriptional regulationaffects soluble Rnr1 protein levels.

RNR1 is transcribed and translated after DNA damageAs a consequence of observingMMS-induced increases in

only 1 of the 2 homologous large subunits of the RNRcomplex (i.e., Rnr3 and not Rnr1), we have focused in-depth studies on the potential posttranscriptional regulationof the RNR1 transcript and the regulation of the Rnr1protein. To further validate our observation that increasedRNR1 transcription does not lead to increased levels ofsoluble protein, we monitored RNR1 and RNR1-TAPtranscript and protein levels in both MMS-treated anduntreated controls (10 to 60 minutes postdamage) usingNorthern and Western blots (Fig. 2A). Rnr1 protein levelswere also analyzed using 2 different antibodies: an anti-TAPantibody to examine the levels of endogenously expressedRnr1-TAP and an anti-Rnr1 antibody to analyze the nativeprotein. We determined that at 10 and 60 minutes afterMMS treatment, RNR1 transcription increased 2.0- to 3.5-fold. Similar MMS-induction results were obtained for theRNR1-TAP transcript. Rnr1 and Rnr1-TAP protein levelsin the soluble fraction showed little increase at any time

point when compared with untreated controls (Fig. 2B).Analogous Northern and Western blot results wereobserved at 20-, 30-, and 45-minute time points (Supple-mentary Fig. S1D). Our time course results using bothendogenous and TAP-tagged Rnr1 further demonstrate thatlevels of soluble Rnr1 protein do not dramatically increaseafter MMS damage. In addition, the parallel behavior of theendogenous and TAP-tagged Rnr1 proteins supports thatthe C-terminal TAP tag on Rnr1 does not promote proteindegradation.We have performed similar time-course studiesusing our quantitative ECL approach to track levels ofendogenous Rnr1, and in all cases we observed minimalchange in soluble Rnr1 protein levels after MMS treatment(data not shown). The observed decoupling of the MMS-induced RNR1 transcriptional increase from a solubleprotein level increase suggests that a fraction of theRNR1 transcript or Rnr1 protein pools were affectedby posttranscriptional, translational, or posttranslationalregulation.Protein levels can be regulated during translation or by

protein degradation pathways. Translation occurs in theribosomes and polysome profiles are typically used toidentify transcripts engaged with the translation machinery.We performed a polysome profile to determine if the RNR1transcript is actively translated after MMS damage(Fig. 2C). Analysis of the profile for RNR1 from theuntreated sample indicated that a portion of the transcriptwas found in the polysomes, supporting active translation ofthe transcript under basal conditions. After MMS treat-ment, we observed that a majority of the RNR1 transcriptswere contained in the polysome portion of our profile,indicating that RNR1 is actively translated after damage.Taken together, our RNR1 polysome profile results foruntreated and MMS-treated cells suggest that Rnr1 isregulated posttranslationally.

Rnr1 protein degradation is vacuole-dependent andincreased after DNA damageWe reasoned that our failure to observe an increase in

soluble Rnr1 protein levels after MMS damage was mostlikely caused by the degradation of newly synthesized orexisting Rnr1 protein. Protein degradation occurs via pro-teasome- or vacuole-dependent pathways, with the laterpathway using membrane-bound structures to remove pro-tein from the soluble fraction found in the cytoplasm anddeliver it to the vacuole. To determine if either proteasome-or vacuole-dependent pathways promoted the degradationof Rnr1 after MMS damage, we analyzed MMS-inducedRnr1 protein levels in the presence of the proteasomeinhibitor MG132 or the vacuolar inhibitor PMSF(14, 40). Preincubation of cells with PMSF followed byMMS treatment promoted a notable increase in the solublelevels of Rnr1 protein at 60 minutes (Fig. 3A), whereas theproteasome inhibitor MG132 did not alter the levels ofobserved protein. We have also analyzed Rnr4 protein levelsafter treatment with PMSF and found that inhibition ofvacuole-dependent degradation did not affect soluble Rnr4protein levels (Supplementary Figure S2A). This negative

Dyavaiah et al.


result suggests that regulation of Rnr4 occurs outsideof vacuole-associated protein degradation pathways. Incontrast, our positive results showing an increase in Rnr1after PMSF treatment lead us to propose that Rnr1 isultimately degraded in a vacuolar-dependent fashion.To explore the potential role of the vacuole in Rnr1

protein degradation, we used a panel of yeast mutantsspecific to a component of the Golgi-associated retrogradeprotein complex involved in vacuolar protein sorting(vps54D) and 5 VMA gene mutants (vma2D, vma4D,vma6D, vma7D, and vma21D). VMA genes code for vacuo-lar Hþ–ATPases that acidify vacuoles and promote proteo-lysis by vacuolar peptidases (41). Rnr1 protein levels weremeasured in untreated and MMS-treated cells for the panelof mutants, as well as for wild-type controls.We observed anapproximately 1.5-fold increase in basal Rnr1 protein levelsin the vma4D, vma6D, vma7D, and vma21D cells relative towild-type cells (Supplementary Fig. S2B), suggesting thatsome vacuolar-mediated degradation of Rnr1 occurs under

normal growth conditions. We observed MMS-inducedincreases in Rnr1 protein levels in the vps54D, vma2D,vma4D, vma6D, vma7D, and vma21D cells, relative to eachmutant-specific untreated control (Figs. 3 B and C). In allcases, the MMS-induced increase of Rnr1 in the vacuole-compromised mutants, relative to untreated mutants, wascontrasted by little MMS-induced change in soluble Rnr1protein levels in wild-type cells after damage (Fig. 3C). Ourresults with the pharmacological inhibitors PMSF andMG132 and data using vacuole-compromised mutantssupport the hypothesis that vacuole-dependent degradationof Rnr1 is occurring.

The Rnr1 protein is actively targeted to the vacuoleafter DNA damageWe reasoned that for vacuole-dependent degradation of

Rnr1 to occur, a portion of the Rnr1 protein pool must berelocalized to the vacuole. It has previously been demon-strated that subcellular relocalization of the RNR small

Figure 2. The MMS-inducedincrease in RNR1 transcriptionand corresponding polysomeoccupancy do not lead toincreased Rnr1 protein levels. A,cultures of wild-type cells(BY4741 or ATCC201388) weredivided equally and either leftuntreated or exposed to 0.0125%MMS for 10 to 60 minutes. TotalRNA and protein were extractedfrom all samples. EndogenousRNR1, RNR1-TAP and ACT1transcripts were analyzed byNorthern blot. Rnr1 and Rnr1-TAPprotein levels were measured byWestern blot using anti-Rnr1 andanti-TAP antibodies. As a loadingcontrol, tubulin levels were alsoanalyzed by Western blot. B, thebar graph represents the foldchange in protein and transcriptlevels after MMS damage intreated samples, as comparedwith untreated samples. Individualbands were quantitated bydensitometry and compared to theloading controls, ACT1 andtubulin. C, cell lysates of MMS-treated (60 minutes) and untreatedcells were separated on a 10% to50% sucrose gradient; RNA wasextracted from each fraction andACT1 and endogenous RNR1were probed by Northern blot.Ethidium bromide staining wasused to indicate the relativeamounts of 28S and 18S rRNA ineach fraction. Note the middlepeak in the profile corresponds tothe monosome.

C

28S18S

ACT1

RNR1

Rel

ativ

e O

D25

4

Untreated MMS-treated

1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Polysome Polysome

Rnr1-TAP

RNR1

ACT1

Tubulin

Rnr1

Tubulin

minutes10 60MMS - + - +

RNR1-TAP

ACT1

Transcript

TAP-tagged protein

Protein

TAP-taggedTranscript

A B

Fol

d ch

ange

(M

MS

/unt

reat

ed)

0

1

2

3

4

RN

R1-

TAP

RN

R1

Rnr

1-TA

P

Rnr

1

RN

R1-

TAP

RN

R1

Rnr

1-TA

P

Rnr

1

10 min.Transcript Protein Transcript Protein

10 minutes 60 minutes



subunits from the nucleus to the cytoplasm occurs, and thissupports the argument that RNR activity can be regulatedby subunit location (42). We have demonstrated that afraction of Rnr1 is degraded in a vacuole-dependent fashion(Fig. 3). We used endogenously expressed Rnr1-GFP andRnr3-GFP to study the location of the large subunits pre-and post-MMS damage. Fluorescence analysis under basalconditions demonstrated that Rnr1 was diffuse and cyto-plasmic with some discrete foci (38% � 7% of total cells, n¼ 3) close to the vacuole in the cell (Fig. 4A and B). Wenote that these foci can also be observed in the Rnr1-specificdata found in the Yeast GFP Fusion Localization Database(yeastgfp.yeastgenome.org; ref. 43). In addition, we notethat even when cells have discrete foci, we also detected adiffuse cytoplasmic signal for Rnr1-GFP. We determinedthat MMS treatment induced the transcription of RNR1and significantly increased the formation of discrete foci(73% � 8%, n ¼ 3; Fig. 4B), thus indicating a relocaliza-tion of a portion of Rnr1-GFP after MMS damage. Incontrast, Rnr3-GFP remained diffuse and predominantlycytoplasmic under basal andMMS-treated conditions (Sup-plementary Fig. S3A). We have also analyzed the growthrate and viability of the Rnr1-GFP strain, before andafter MMS treatment and demonstrated that it behaves

like wild-type cells (Supplementary Fig. S3B–C). If the GFPtag destabilized the essential protein Rnr1, growth rate andMMS phenotypes would be expected, and this was notobserved. Further, data from half-life experiments indicatesthat both native Rnr1 and Rnr1-GFP are long-lived proteinswith half-lives greater than 8 hours (SupplementaryFig. S3D). Together, these results support that the C-terminal GFP tag on Rnr1 is not hindering RNR activityor destabilizing the Rnr1 protein. Further they demonstratethat the Rnr1 protein is long lived under basal conditions.We note that for Rnr1-GFP, DNA damage appeared toinduce an increase in focus size, a finding that we confirmedusing quantitative image analysis. When compared with theuntreated cells (478 � 290 arbitrary units), the focus areawas increased approximately 2.2-fold in MMS-treated cells(1,030 � 478 arbitrary units). The increase in the numberand area of Rnr1-GFP foci in the MMS-treated cells,relative to untreated cells, reveals that a large fraction ofthe Rnr1 protein pool is localized to foci following DNAdamage.To further analyze the location of Rnr1-GFP foci, we

used fluorescence and confocal microscopy with 2 differ-ent vacuole-specific dyes: FM4-64 stains the vacuolarmembrane and CellTracker Blue selectively stains thevacuolar lumen. Rnr1-GFP–expressing cells were treatedwith MMS and counterstained with each of the vacuolemarkers. Merged images specific to Rnr1-GFP foci andthe vacuole membrane stained with FM4-64 indicatedthat the Rnr1-GFP foci were docked at the vacuolemembrane and poised to enter the vacuole (Fig. 4C).We obtained a similar result using CellTracker Blue: theRnr1-GFP foci were observed overlapping the peripheryof the vacuole (Fig. 4D). The fluorescence and confocalmicroscopy results support the hypothesis that Rnr1-GFPfoci are targeted to the vacuole and that Rnr1 degradationoccurs in the vacuole.Our observations from experiments using both native

Rnr1 and a GFP-tagged version predicted that a portion ofthe Rnr1 protein pool was being packaged into a mem-brane-bound vesicle and targeted for degradation. To testthis prediction, we isolated the supernatant and membranefraction of cells from mock- and MMS-treated cells. Asexpected the levels of native Rnr1 in the supernatant fromuntreated and MMS-treated cells was similar. In contrast,we have determined that there is an approximately 2-foldincrease in native Rnr1 in the membrane fraction afterMMS damage (P < 0.0135), relative to the amount of Rnr1found in the membrane fraction of untreated cells (Figs. 4Eand F). This finding supports that Rnr1 is transported to thevacuole in membrane-bound autophagosomes for degrada-tion. RNR activity is regulated during the cell cycle. Wereasoned that the removal Rnr1 from the soluble fractioncould help optimize dNTP levels or promote cell-cycletransitions and in either case Rnr1 targeting to the autop-hagosome should be cell-cycle–dependent. In addition, theautophagosome results, when combined with previous data,predict that Rnr1 foci formation and transport to thevacuole should also be dependent on autophagy and TOR.

B

C

WT vps54Δ vma2Δ vma4Δ vma6Δ vma7Δ vma21Δ- + - + - + - + - + - + - +

Rnr1

Tubulin

MMS

A

0

0.5

1

1.5

2

2.5

WT vps54Δ vma2Δ vma4Δ vma6Δ vma7Δ vma21Δ

Fol

d ch

ange

(M

MS

/unt

reat

ed)

Rnr1

Tubulin

---

MMSPMSF

MG132--

++-

+-+

+

Figure 3.Compromised vacuolar function promotes increased Rnr1 levelsafter MMS treatment. A, S. cerevisiae (BY4741) erg6D cells were grown at30� C to 5 � 106 cells/mL in YPD medium. The cells deficient in Erg6 areconsidered leaky and were used to facilitate entry of the inhibitors. Aculture of cells was divided equally and preincubated with either 1 mmol/LPMSF or 100 mmol/L MG132 for 90 minutes, and then treated with MMS(0.0125%) for 60 minutes. B, Rnr1 protein levels in wild-type and vacuolarmutants were measured by Western blot before and after 0.0125% MMSexposure, using an Rnr1-specific antibody. C, quantification of Rnr1protein levels demonstrated that there is an MMS-induced increase insoluble Rnr1 protein in vacuolar mutants (n ¼ 3), when compared withcorresponding untreated cells; this is contrasted in wild-type cells by littlechange in soluble Rnr1 protein levels after MMS treatment.

Dyavaiah et al.


Rnr1-GFP foci levels peak during late S and early G2 ofthe cell cycleMMS promotes an S-phase arrest and induces the DNA

damage response. During this response, cells will optimizeenzyme activities to promote efficient DNA replication. Wereasoned that if Rnr1 targeting to the vacuole is used tooptimize or control DNA synthesis, then Rnr1-GFP foci

should increase during S-phase. To test this prediction, cellswere synchronized, released from G1, and analyzed forRnr1-GFP foci formation as a function of time (0, 15,30, 45, 60, 90, and 120 minutes). Cell-cycle characteristicswere also quantitated by FACS analysis at each of the 7 timepoints. We observed (Fig. 5A) foci at increasing levels aftercells were released from G1. Peak foci formation occurred at

D

Bright field Rnr1-GFP FM4-64 Merge

Bright field Rnr1-GFP Cell tracker blue Merge

C

Bright field Rnr1-GFP Merge

MM

S

A

FE

MMS - + - +

Rnr1

Tubulin

Supernatant Membrane extract

Rel

ativ

e R

nr1

prot

ein

leve

l

P = 0.0135

B

P < 0.005U

ntre

ated

100

80

60

40

20

0

Num

ber

of c

ells

with

foci

Unt. MMS

1.41.2

10.80.60.40.2

0Untreated

Supernatant Membrane extract

MMS Untreated MMS

Figure 4. Increased Rnr1-GFP localization to the vacuole after MMS damage. A, Rnr1-GFP expression in untreated and MMS-treated cells (n ¼ 100) wasexamined by bright field and fluorescencemicroscopy. B, foci were quantitated and the number of cells with foci (P < 0.05) was significantly increased inMMS-treated cells. C, cells were stained with FM4-64 to localize the vacuolar membrane; images for green fluorescence, red fluorescence, and bright fieldmicroscopy were merged as indicated. D, cells that were stained with Cell Tracker Blue, to localize the vacuolar lumen, were observed in green and DAPIchannels, and then merged. E, Rnr1 levels in wild-type cells left untreated or exposed to 0.0125%MMS were measured as described in Figure 3. Rnr1 proteinlevels were analyzed in the supernatant and in the protein isolated from the membrane fraction. F, Rnr1 protein levels were normalized to the loading controltubulin and relative levels were compared using Student's t test. As expected, tubulin levels were different in the supernatant andmembrane extract. This data,therefore, should not be used to compare absolute levels between the supernatant and membrane fractions.



60 minutes (Fig. 5B), a time that coincides with a majorityof the cells being in late S or early G2 phases, withapproximately half in each phase. As this is a populationof cells, this point represents the S to G2 transition point.Peak mRNA expression for RNR1 occurs during S, whereasthe foci were present during both S and early G2; this mostlikely occurs because of the high translation level of RNR1throughout S-phase and its increased targeting to thevacuole during late S and into early G2. Our results usingsynchronized and released cells confirm our prediction thatfoci formation is cell-cycle dependent.

Autophagy-associated proteins target Rnr1 to thevacuole for degradationTargeting macromolecules or organelles to the vacuole

requires Atg proteins that participate in autophagy. Yeast atgmutants are defective in autophagy and have been reportedto be corrupted in vacuole-dependent protein degradation(44). Based on our Western blot and microscopy results, wepostulated that Rnr1 degradation would be dependent on

Atg proteins. To test this hypothesis, we focused our studyon native Rnr1 protein levels in atg2D, atg5D, atg8D, atg9D,and atg14D cells. Cultures of atg mutants were either leftuntreated or treated with 0.0125% MMS for 1 hour, afterwhich Rnr1 protein levels were measured by Western blot.Relative to each untreated mutant, we observed MMS-induced increases in Rnr1 protein levels in our panel of atgmutants (Fig. 6A), with 2.0- to 3.5-fold increases for Rnr1identified in atg2D, atg5D, atg8D, atg9D, and atg14D(Fig. 6B). In contrast, wild-type cells demonstrated littlechange in soluble Rnr1 protein levels after MMS damage.Atg2, Atg5, Atg8, Atg9, and Atg14 are proteins involved inautophagosome assembly or vacuolar docking, and defi-ciencies in any of these proteins increased soluble Rnr1protein levels. Western blot results reveal that proteinsparticipating in autophagy are involved in the removal ofRnr1 from the soluble fraction. In addition, they demon-strate that deficiencies in either autophagosome assembly orvacuolar docking result in increased levels of soluble Rnr1protein after MMS damage.

A

B

PI-API-A PI-A PI-A

400

300

200

100

0

Cou

nts

Rnr

1-G

FP

α factor synchronized

YPD released

Untreated MMS

Time (minutes)

Per

cent

age

of c

ells

Foci

S

G2

70

60

50

40

30

20

10

0

0 15 30 45 60 90 120

Figure 5. Rnr1-GFP foci increaseduring S and peaks at the S-G2

border. A, yeast Rnr1-GFPcultures were grown at 30�C to5 � 106 cells/mL in YPD, treatedwith alpha factor, and thenreleased into YPD for 60 minutes.Samples were analyzed by flowcytometry and florescencemicroscopy to quantitate cell-cycle progression and fociformation. For reference,asynchronous cells were treatedwith MMS for 60 minutes andanalyzed as described previously.B, quantitative analysis of cell-cycle stage and Rnr1-GFP focilevels in cells released from G1

(n ¼ 2) was performed asdescribed previously. Green, blue,and red bars represent thepercentage of cells with foci, inS-phase or in G2-phase,respectively.

Dyavaiah et al.


Four lines of evidence support that Rnr1 is sequesteredinto autophagosomes and subsequently released into thevacuole for degradation. Included among these were theobservations that Rnr1-GFP foci localized at the vacuoleand that Rnr1 protein levels were increased in membranefractions after MMS treatment, relative to untreated cells(Fig. 4E and F). In addition, we also observed MMSinduced increases in soluble Rnr1 protein levels invacuole-compromised and atg mutants (Fig. 3B and Cand 6A and B). Consequently, we postulated that defectsin the early stages of the autophagy pathway would preventthe formation of autophagosomes, which are represented asdiscrete Rnr1-GFP foci. We deleted specific atg genes(atg2D, atg5D, atg8D, atg9D, and atg14D) in the Rnr1-GFP strain and analyzed Rnr1-GFP levels and their corre-sponding foci, both before (data not shown) and afterMMS-exposure (Fig. 6C). Rnr1-GFP was found through-out the cytoplasm before and after MMS exposure in atg2D,atg5D, atg8D, atg9D, and atg14D; in contrast, we observedMMS-induced foci in wild-type cells. We note that inFigure 6, experiments with atg mutants were performedusing both native Rnr1 (Fig. 6A and B) and a C-terminalGFP-tagged version (Fig. 6C), and results associated witheach form of Rnr1 implicate a role for autophagy proteins inregulating soluble protein levels.

TOR inhibition and activation of autophagy promotesRnr1 degradationAutophagy is increased when TOR activity is inhibited by

nutrient starvation or rapamycin. We reasoned that if Rnr1

is degraded via autophagy, then Rnr1 protein levels shouldbe sensitive to TOR activity. Rnr1 protein and RNR1transcript levels were analyzed in wild-type cells, both beforeand after treatment with rapamycin for 60 minutes(Fig. 7A). Although the RNR1 transcript was detected underall conditions and induced after MMS treatment, there wasa large decrease in Rnr1 protein levels in cells treated withrapamycin when compared with those left untreated. Nota-bly, RNR1 transcript levels in the rapamycin þ MMSsamples are higher than in the untreated sample, yet theRnr1 protein levels are much lower in this treated sample.Further, we found that this rapamycin-induced decrease inRnr1 protein levels are diminished in cells corrupted inautophagy (atg9D; Fig. 7B). The rapamycin-induceddecrease in Rnr1 protein levels suggest that foci formquickly in response to treatment. To test our predictionwe analyzed Rnr1-GFP foci levels 15 minutes after cells hadbeen subjected to rapamycin treatment (Fig. 7 C and D).We found a significant increase (�35%, P < 0.007) inRnr1-GFP foci levels in rapamycin-treated cells, whencompared with those left untreated. In addition, we ana-lyzed native Rnr1 protein and transcript levels in cells whereTOR was inhibited by nutritional stress (Fig. 7E). Wefound that similar to our results with rapamycin, nutritionalstress promoted a substantial decrease in soluble Rnr1protein levels in wild-type cells. This decrease in Rnr1protein levels was abrogated in the autophagy mutant'satg9D and atg14D. Our observation of high Rnr1 proteinlevels in atg9D and atg14D cells, relative to wild-type cells,was a striking result because the RNR1 transcripts were at

Figure 6. Autophagy mutantspromote increased Rnr1 levelsand mislocalization of Rnr1. A,Rnr1 and tubulin levels in wild-type and autophagymutants, fromboth untreated and MMS-treatedcells, weremeasured as describedpreviously. Fifty micrograms oftotal protein was loaded for eachsample. B, Rnr1 protein levels inMMS and untreated cells werequantitated relative to tubulin, andthese values were used togenerate fold changes after MMStreatment. C, wild-type andautophagy deletion mutants(atg2D, atg5D, atg8D, atg9D, andatg14D) endogenously expressingRnr1-GFP were grown to 5 � 106

cells/mL in YPD and exposed to0.0125% MMS for 1 hour. Cellswere observed under thefluorescence microscope both inbright field and greenfluorescence mode, after whichthe images were merged.

MMS - + - + - + - +

Rnr1

WT atg2Δ atg5Δ atg8Δ

MMS - + - + - +

Rnr1

WT atg9Δ atg14Δ

Tubulin

Tubulin

A

WT atg2Δ atg5Δ atg8Δ atg9Δ atg14Δ

WT

Bright Field Rnr1-GFP Merge

atg2Δ

atg5Δ

atg8Δ

atg9Δ

atg14Δ

C

B 5

4.5

Fol

d ch

ang

(MM

S/u

ntre

ated

)

4

3.5

3

2.5

2

1.5

1

0.5

0



higher levels in wild-type cells. Ultimately, these resultsprovide evidence that TOR inhibition and the induction ofautophagy will promote the removal of Rnr1 from thesoluble protein fraction.In wild-type cells, the degradation of Rnr1 after TOR

inhibition most likely prevents DNA synthesis after DNAdamage, and we reasoned this should promote an efficientresponse to DNA damage. Conversely, we postulated that afailure to degrade Rnr1 levels after TOR inhibition shouldlead to a DNA damage phenotype in atg mutants. To testour prediction, we exposed cells to rapamycin in the absenceor presence of a DNA damaging agent (Fig. 7F). We usedrapamycin in plate-based experiments to induce autophagy,as atg mutants can grow in the presence of this compoundbut fail to grow on nutritional stress media. Thus, the use ofrapamycin provided us with a methodology to test for aDDR-associated growth phenotype in cells that fail to

degrade Rnr1 via autophagy. Conditions were chosen soneither wild-type or atg cells exhibited a growth defect in thepresence of rapamycin or MMS alone. Plate-based killingassays using rapamycin demonstrate that atg (atg2D, atg5D,atg8D, atg9D, atg12D, and atg14D) mutants are moresensitive to MMS, relative to wild-type cells. Takentogether, our results support the idea that TOR inhibitionand the activation of autophagy plays an important role inpromoting the degradation of Rnr1, and our phenotypicstudies demonstrate that defects in this process lead todefects in the DNA damage response.

Discussion

Targeted gene- and protein-specific studies report proteinregulation at the levels of protein synthesis or proteindegradation (15, 45); yet, few global techniques have been

ARAP+

UNT MMS RAP MMS

Rnr1

Tubulin

RNR1

ACT1

Bright Field Rnr1-GFP Merge

UNT

RAP

C

D

- + - + - +WT atg9Δ atg14Δ

Nutritional stressE

Rnr1

Tubulin

RNR1

ACT1

MMS

WT atg9Δ

Rnr1

Tubulin

B- + - +RAP

Num

ber

of c

ells

with

foc

i P < 0.007

FWT

atg2Δ

atg8Δ

atg9Δ

atg5Δ

atg12Δ

atg14

YPD MMS RAP WT

atg2Δ

atg8Δ

atg9Δ

atg5Δ

atg12Δ

atg14Δ

YPD MMS RAP MMS + RAP

100

80

60

40

20

0

WT RAP

Figure 7. Autophagy-dependentdegradation of Rnr1 after TORinhibition. A, wild-type cells wereexposed for 60 minutes to either0.0125% MMS, 50 ng/mLrapamycin, or both. Native Rnr1and tubulin protein levels, as wellas RNR1 and ACT1 transcriptlevels, from untreated and treatedcells were measured as describedin Figure 2B, wild-type and atg9Dcells were grown in YPD andtreated with rapamycin for 45minutes. Native Rnr1 and tubulinlevels from untreated and treatedcells were measured as describedpreviously. C, Rnr1-GFPexpression in cells left untreated(UNT) or treated with rapamycinfor 15 minutes was examined bybright field and fluorescencemicroscopy. D, foci in C werequantitated and out of 100 cellsthe number of cells with Rnr1-GFPfoci (P < 0.007), was significantlyincreased in rapamycin-treatedcells. E, wild-type, atg9D, andatg14D cells were grown in YPD to5 � 106 cells/mL and then furthergrown in starvation media for 12hours, followed by treatment with0.0125% MMS for 1 hour. NativeRnr1 and tubulin protein levels, aswell as RNR1 and ACT1 transcriptlevels, from untreated and treatedcells were measured as describedin Figure 2F, overnight cultures ofcells were serially diluted andplated on YPD or YPDsupplemented with 0.0125%MMS, 25 ng/mL rapamycin (RAP),or 0.0125% MMS and 25 ng/mLRAP.

Dyavaiah et al.


developed to efficiently identify multiple targets of post-transcriptional regulation. Our developed methodologyuses 2 standard high-throughput approaches tethered tocomputational analysis to systematically identify targets ofposttranscriptional regulation. Our approach has the poten-tial to identify a plethora of novel regulatory strategiesbecause it can be applied to other perturbations and modelsystems. What we have demonstrated is that matchedtranscript-protein level studies can be used to identifydiscrepancies in standard gene regulatory models (i.e.,transcript is induced to make more protein). The identifica-tion of any discrepancy or unusual regulatory pattern can bechallenging to comprehend but detailed study ultimatelyleads to a better understanding of biology. Here we focusedour studies on the large subunit of the RNR complex toelucidate a novel pathway for the posttranslational regula-tion of Rnr1 protein levels.Because dNTP levels play an important role in check-

point arrest, mutagenesis, and cell viability after DNAdamage, RNR activity is extensively regulated in eukaryoticorganisms (3). In mammals, RNR activity is regulated by S-phase transcription, the p53-inducible subunit p53R2,proteasome-dependent degradation of the small subunitR2, and allosteric mechanisms (46). In yeast, transcriptionalactivation of the subunits comprising the RNR complexrequires derepression of Rfx1 by Dun1-dependent phos-phorylation (47), with phosphorylation also being used toinactivate the RNR inhibitor Sml1. Recently, translation ofthe RNR1 and RNR3 transcripts has been reported to bedependent on Trm9-catalyzed tRNA modifications (48)and our current study demonstrates that the soluble levels ofthe large Rnr1 subunit can be regulated via autophagy.Protein degradation of RNR subunits has now beenobserved in both yeast and human systems, thus supportingthe hypothesis that subunit degradation is a conservedregulatory mechanism controlling dNTP levels in vivo.In general, proteasome-dependent degradation of specific

proteins uses ubiquitination and is associated with theremoval of short-lived proteins, whereas vacuolar-depen-dent degradation is considered a nonspecific mechanism toremove long-lived proteins via autophagy. Our results runcontrary to this dogma: we observed a very specific removalof Rnr1 from the soluble protein pool by autophagosometransport to the vacuole. Although autophagy is generallyconsidered to be a bulk degradation process, examples ofspecific targeting have been reported for the cytosoliccysteine protease Lap3 (49), the gluconeogenic enzymefructose-1,6-bisphosphatase (FBPase; ref. 50), and themetabolic enzyme acetaldehyde dehydrogenase (Ald6; ref.51). Overexpression of Ald6 has been shown to be cyto-toxic, implying that the targeted degradation of this proteinis advantageous. Removal of proteins that produce domi-nant-negative phenotypes may be a theme associated withthe specific targeting of proteins via autophagy.A major question arising from our findings is: Why do

cells transcribe and translate Rnr1 when it is unnecessary? Inanswering this question, we note that our results detailingRnr1 protein levels involved 2 distinct yet overlapping

responses: DNA damage and nutrient stress (via TORinhibition). In theory, each response has a different objec-tive and thus a different endpoint for Rnr1 protein levels.The DNA damage response requires a precise amount ofRnr1 to promote DNA synthesis under damaging condi-tions and to control cell-cycle transitions. The degradationof excess Rnr1 protein during DNA damage may berequired to promote the formation of an Rnr1–Rnr3assembly in the final RNR complex, as opposed to anRnr1–Rnr1 assembly. The Rnr1–Rnr3 assembly is reportedto be the most catalytically active form (52). Ghaemma-ghami and colleagues have reported that Rnr1 protein levelsare approximately 200-fold higher than Rnr3 levels (53).Thus, a low level of Rnr1 protein localization to autophago-somes, as seen in our cell-cycle data for early to mid S-phase,and its subsequent degradation may be part of the mechan-ism driving subunit formation and optimizing catalyticactivity of the RNR tetramer. Because high dNTP levelsare needed to drive translesion and replicative polymerasesduring the DNA damage response, optimal RNR activity isessential to cellular proliferation (3, 4, 54). In addition, ourcell-cycle data demonstrating peak Rnr1-GFP foci at the S–G2 border suggests that increased Rnr1 degradation couldalso be used to facilitate transition into G2. There is adecrease in dNTP levels during G2, relative to S, and Rnr1degradation could be one of the mechanisms used topromote this reduction. The autophagy-dependent degra-dation of Rnr1 could play multiple roles, ultimately webelieve at low levels it acts to optimize RNR activity duringS-phase and at high levels it promotes a transition to G2.Nutrient stress, which is also mimicked by rapamycin

treatment, is the other distinct condition used in our studyand promotes the removal of all Rnr1 protein. The Rnr1protein was not detected 60 minutes after TOR inhibition,and its absence should lead to a decrease in dNTP levels anda halt to DNA synthesis. Increased autophagy drives thisdegradation of Rnr1, and that may be a mechanism toquickly align DNA synthesis with current nutrient condi-tions. TOR inhibition has also been demonstrated to lead toa reduction in RNR1 transcription (55), and we haveobserved this transcriptional control in our studies. Ourreport supports and demonstrates that the cell uses 2mechanisms to decrease Rnr1 protein levels after TORinhibition, an RNR1 transcriptional decrease and a removalof existing Rnr1 from the soluble pool, followed by degra-dation. We note that we have observed a long half life forRnr1 protein. The 2 TOR-based strategies to control Rnr1protein levels limit the synthesis of new protein and ensurethat any existing protein is removed from the solublefraction. In situations of nutrient stress, cell divisiondecreases and autophagy ensures that the necessary pre-cursors are available to maintain the cell. In this context,Rnr1 degradation might represent one of the controlmechanisms used to arrest cells and prevent DNA synthesisduring times of nutritional stress. Our phenotypic studiesdemonstrate the importance of coordinated response toDNA damage under conditions of TOR inhibition, asatgmutants fail to grow under these conditions. Ultimately,



our report links defects in autophagy to a DNA damagephenotype and highlights the importance of protein degra-dation for optimizing the DNA damage response.In conclusion, we have used systems biology and targeted

studies to identify posttranscriptional regulatory targets andto demonstrate that autophagy plays an important role inregulating soluble Rnr1 protein levels. A remaining questiondeals with how Rnr1 is targeted by the autophagy machin-ery. It is possible that MMS-induced protein damage,aggregation, or a posttranslational modification providesthe signal, but this has yet to be determined and is the focusof future work. Previous studies in human cell systemsreport that defects in autophagy promote genome instabil-ity, arising from increased metabolic stress that generatesincreased levels of endogenous damaging agents (56). Inaddition, mTOR inhibition and the induction of autophagyhave been linked to p53 and Ataxia-telangiectasia mutated(ATM) signaling after DNA damage, with ATM activationalso occurring by reactive oxygen species in the cytoplasm(57, 58). Nonetheless, these aforementioned examples alldemonstrate links between autophagy and known compo-nents of the DNA damage response. Our findings in yeast

further demonstrate that autophagy and DNA damagesignaling are intricately coordinated, and lead us to spec-ulate that human cells defective in autophagy may havealtered levels of DNA damage response proteins. Coupledwith the increase in DNA damage, a misregulation of DNAdamage response proteins could dramatically increase gen-ome instability and carcinogenesis in autophagy deficientcells.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work was supported by grants awarded to T.J. Begley from theNIH (ES01225101 and ES015037) and a James D. Watson Award throughNYSTAR.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement in accor-dance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received October 21, 2010; revised December 10, 2010; accepted December 14,2010; published OnlineFirst February 22, 2011.

References1. Gardner R, Putnam CW, Weinert T. RAD53, DUN1 and PDS1 define

two parallel G2/M checkpoint pathways in budding yeast. Embo J1999;18:3173–85.

2. Zhao X, Rothstein R. The Dun1 checkpoint kinase phosphorylates andregulates the ribonucleotide reductase inhibitor Sml1. Proc Natl AcadSci U S A 2002;99:3746–51.

3. Chabes A, Georgieva B, Domkin V, Zhao X, Rothstein R, Thelander L.Survival of DNA damage in yeast directly depends on increased dNTPlevels allowed by relaxed feedback inhibition of ribonucleotide reduc-tase. Cell 2003;112:391–401.

4. Chabes A, Stillman B. Constitutively high dNTP concentration inhibitscell cycle progression and the DNA damage checkpoint in yeastSaccharomyces cerevisiae. Proc Natl Acad Sci U S A 2007;104:1183–8.

5. Vallen EA, Cross FR. Interaction between the MEC1-dependent DNAsynthesis checkpoint and G1 cyclin function in Saccharomyces cer-evisiae. Genetics 1999;151:459–71.

6. Zhao X, Chabes A, Domkin V, Thelander L, Rothstein R. The ribonu-cleotide reductase inhibitor Sml1 is a new target of the Mec1/Rad53kinase cascade during growth and in response to DNA damage. EmboJ 2001;20:3544–53.

7. Gasch AP, Spellman PT, Kao CM, Carmel-Harel O, EisenMB, Storz G,et al. Genomic expression programs in the response of yeast cells toenvironmental changes. Mol Biol Cell 2000;11:4241–57.

8. Gasch AP, Werner-Washburne M. The genomics of yeast responsesto environmental stress and starvation. Funct Integr Genomics2002;2:181–92.

9. Jelinsky SA, Estep P, Church GM, Samson LD. Regulatory networksrevealed by transcriptional profiling of damaged Saccharomycescerevisiae cells: Rpn4 links base excision repair with proteasomes.Mol Cell Biol 2000;20:8157–67.

10. Jelinsky SA, Samson LD. Global response of Saccharomyces cerevi-siae to an alkylating agent. Proc Natl Acad Sci U SA 1999;96:1486–91.

11. Workman CT, Mak HC, McCuine S, Tagne JB, Agarwal M, Ozier O,et al. A systems approach to mapping DNA damage response path-ways. Science 2006;312:1054–59.

12. Hall J, Montesano R. Alkylation and oxidative-DNA damage repairactivity in blood leukocytes of smokers and non-smokers. Interna-tional J Cancer 1993;54:728–33.

13. Hanway D, Chin JK, Xia G, Oshiro G, Winzeler EA, Romesberg FE.Previously uncharacterized genes in the UV- and MMS-induced DNAdamage response in yeast. Proc Natl Acad Sci U S A 2002;99:10605–610.

14. Burgis NE, Samson LD. The Protein Degradation Response of Sac-charomyces cerevisiae to Classical DNA-Damaging Agents. ChemRes Toxicol 2007.

15. Hall JR, Kow E, Nevis KR, Lu CK, Luce KS, Zhong Q, et al. Cdc6stability is regulated by the Huwe1 ubiquitin ligase after DNA damage.Mol Biol Cell 2007;18:3340–50.

16. Kitanovic A, Wolfl S. Fructose-1,6-bisphosphatase mediates cellularresponses to DNA damage and aging in Saccharomyces cerevisiae.Mutat Res 2006;594:135–47.

17. Begley TJ, Rosenbach AS, Ideker T, Samson LD. Hot spots formodulating toxicity identified by genomic phenotyping and localiza-tion mapping. Mol Cell 2004;16:117–25.

18. Said MR, Begley TJ, Oppenheim AV, Lauffenburger DA, Samson LD.Global network analysis of phenotypic effects: protein networks andtoxicity modulation in Saccharomyces cerevisiae. Proc Natl Acad SciU S A 2004;101:18006–11.

19. Aouida M, Page N, Leduc A, Peter M, Ramotar D. A genome-widescreen in Saccharomyces cerevisiae reveals altered transport as amechanism of resistance to the anticancer drug bleomycin. CancerRes 2004;64:1102–9.

20. Teter SA, Eggerton KP, Scott SV, Kim J, Fischer AM, Klionsky DJ.Degradation of lipid vesicles in the yeast vacuole requiresfunction of Cvt17, a putative lipase. J Biol Chem 2001;276:2083–7.

21. Nair U, Klionsky DJ. Molecular mechanisms and regulation of specificand nonspecific autophagy pathways in yeast. J Biol Chem 2005;280:41785–88.

22. Kamada Y, Sekito T, Ohsumi Y. Autophagy in yeast: a TOR-mediatedresponse to nutrient starvation. Curr Top Microbiol Immunol 2004;279:73–84.

23. Noda T, Ohsumi Y. Tor, a phosphatidylinositol kinase homologue,controls autophagy in yeast. J Biol Chem 1998;273:3963–6.

24. Guertin DA, Sabatini DM. Defining the role of mTOR in cancer. CancerCell 2007;12:9–22.

Dyavaiah et al.


25. Wong KK. Oral-specific chemical carcinogenesis in mice: an excitingmodel for cancer prevention and therapy. Cancer Prev Res (Phila Pa)2009;2:10–13.

26. Liang XH, Jackson S, SeamanM, Brown K, Kempkes B, Hibshoosh H,et al. Induction of autophagy and inhibition of tumorigenesis by beclin1. Nature 1999;402:672–6.

27. Yu L, Wan F, Dutta S, Welsh S, Liu Z, Freundt E, et al. Autophagicprogrammed cell death by selective catalase degradation. Proc NatlAcad Sci U S A 2006;103:4952–7.

28. Yue Z, Jin S, YangC, Levine AJ, Heintz N. Beclin 1, an autophagy geneessential for early embryonic development, is a haploinsufficienttumor suppressor. Proc Natl Acad Sci U S A 2003;100:15077–82.

29. Karantza-Wadsworth V, Patel S, Kravchuk O, Chen G, Mathew R, JinS, et al. Autophagy mitigates metabolic stress and genome damage inmammary tumorigenesis. Genes Dev 2007;21:1621–35.

30. Jelinsky SA, Estep P, Church GM, Samson LD. Regulatory networksrevealed by transcriptional profiling of damaged saccharomycescerevisiae cells: rpn4 links base excision repair with proteasomes.Mol Cell Biol 2000;20:8157–67.

31. Baldi P, Long AD. A Bayesian framework for the analysis of microarrayexpression data: regularized t -test and statistical inferences of genechanges. Bioinformatics 2001;17:509–19.

32. Arava Y, Wang Y, Storey JD, Liu CL, Brown PO, Herschlag D.Genome-wide analysis of mRNA translation profiles in Saccharo-myces cerevisiae. Proc Natl Acad Sci U S A 2003;100:3889–94.

33. Ghaemmaghami S, Huh WK, Bower K, Howson RW, Belle A,Dephoure N, O'Shea EK, Weissman JS, et al. Global analysis ofprotein expression in yeast. Nature 2003;425:737–741.

34. Luo J, Ning T, Sun Y, Zhu J, Zhu Y, Lin Q, et al. Proteomic analysisof rice endosperm cells in response to expression of hGM-CSF.J Proteome Res 2009;8:829–37.

35. Ross PL, Huang YN, Marchese JN, Williamson B, Parker K, Hattan S,et al. Multiplexed protein quantitation in Saccharomyces cerevisiaeusing amine-reactive isobaric tagging reagents. Mol Cell Proteomics2004;3:1154–69.

36. Pabla R, Pawar V, Zhang H, Siede W, Judith L. Campbell PMCharacterization of Checkpoint Responses to DNA Damage in Sac-charomyces cerevisiae: Basic Protocols. In:Methods in Enzymology,Vol. 409. Academic Press; San Diego, CA 2006. p. 101.

37. Ahnesorg P, Jackson SP. The non-homologous end-joining proteinNej1p is a target of the DNA damage checkpoint. DNA Repair (Amst)2007;6:190–201.

38. Reid RJ, Fiorani P, Sugawara M, Bjornsti MA. CDC45 and DPB11 arerequired for processive DNA replication and resistance to DNA topoi-somerase I-mediated DNA damage. Proc Natl Acad Sci U S A1999;96:11440–5.

39. van Waardenburg RC, Duda DM, Lancaster CS, Schulman BA,Bjornsti MA. Distinct functional domains of Ubc9 dictate cellsurvival and resistance to genotoxic stress. Mol Cell Biol 2006;26:4958–69.

40. Lee DH, Sherman MY, Goldberg AL. Involvement of the molecularchaperone Ydj1 in the ubiquitin-dependent degradation of short-livedand abnormal proteins in Saccharomyces cerevisiae. Mol Cell Biol1996;16:4773–81.

41. Nakamura N, Matsuura A, Wada Y, Ohsumi Y. Acidification ofvacuoles is required for autophagic degradation in the yeast, Sac-charomyces cerevisiae. J Biochem 1997;121:338–44.

42. Yao R, Zhang Z, An X, Bucci B, Perlstein DL, Stubbe J, et al. Sub-cellular localization of yeast ribonucleotide reductase regulated by theDNA replication and damage checkpoint pathways. Proc Natl AcadSci U S A 2003;100:6628–33.

43. Huh WK, Falvo JV, Gerke LC, Carroll AS, Howson RW, Weissman JS,et al. Global analysis of protein localization in budding yeast. Nature2003;425:686–91.

44. Onodera J, Ohsumi Y. Ald6p is a preferred target for autophagy inyeast, Saccharomyces cerevisiae. J Biol Chem 2004;279:16071–6.

45. Lee MW, Kim BJ, Choi HK, Ryu MJ, Kim SB, Kang KM, et al. Globalprotein expression profiling of budding yeast in response to DNAdamage. Yeast 2007;24:145–54.

46. Chabes AL, Pfleger CM, Kirschner MW, Thelander L. Mouse ribonu-cleotide reductase R2 protein: a new target for anaphase-promotingcomplex-Cdh1-mediated proteolysis. Proc Natl Acad Sci U S A2003;100:3925–9.

47. Huang M, Zhou Z, Elledge SJ. The DNA replication and damagecheckpoint pathways induce transcription by inhibition of the Crt1repressor. Cell 1998;94:595–605.

48. Begley U, DyavaiahM, Patil A, Rooney JP, Direnzo D, Young CM, et al.Trm9-Catalyzed tRNA Modifications Link Translation to the DNADamage Response. Mol Cell 2007;28:860–70.

49. Kageyama T, Suzuki K, Ohsumi Y. Lap3 is a selective target ofautophagy in yeast, Saccharomyces cerevisiae. Biochem BiophysRes Commun 2009;378:551–7.

50. Chiang HL, Schekman R. Regulated import and degradation of acytosolic protein in the yeast vacuole. Nature 1991;350:313–8.

51. O’Rourke JP, Hutt JA, DeWille J. Transcriptional regulation of C/EBPdelta in G(0) growth-arrested mouse mammary epithelial cells.Biochem Biophys Res Commun 1999;262:696–701.

52. Domkin V, Thelander L, Chabes A. Yeast DNA damage-inducibleRnr3 has a very low catalytic activity strongly stimulated after theformation of a cross-talking Rnr1/Rnr3 complex. J Biol Chem2002;277:18574–8.

53. Ghaemmaghami S, Huh WK, Bower K, Howson RW, Belle A,Dephoure N, et al. Global analysis of protein expression in yeast.Nature 2003;425:737–41.

54. Sabouri N, Viberg J, Kumar Goyal D, Johansson E, Chabes A.Evidence for Lesion bypass by replicative DNA polymerases duringDNA damage. Nucleic Acids Res 2008;36:5660–7.

55. Shen C, Lancaster CS, Shi B, Guo H, Thimmaiah P, Bjornsti MA. TORsignaling is a determinant of cell survival in response to DNA damage.Mol Cell Biol 2007;27:7007–17.

56. Elliott A, Reiners JJ Jr. Suppression of autophagy enhances thecytotoxicity of the DNA-damaging aromatic amine p-anilinoaniline.Toxicol Appl Pharmacol 2008;232:169–79.

57. Budanov AV, Karin M. p53 target genes sestrin1 and sestrin2 connectgenotoxic stress and mTOR signaling. Cell 2008;134:451–60.

58. Alexander A, Cai SL, Kim J, Nanez A, Sahin M, MacLean KH, et al.ATM signals to TSC2 in the cytoplasm to regulate mTORC1 inresponse to ROS. Proc Natl Acad Sci U S A 2010;107:4153–8.



autophagy-dependent regulation of the dna damage …dna damage and cellular stress responses...

Documents