Molecular Cell

Article

RNA Polymerase I Stability CouplesCellular Growth to Metal AvailabilityYueh-Jung Lee,1 Chrissie Young Lee,1,3 Agnieszka Grzechnik,1,4 Fernando Gonzales-Zubiate,1,5 Ajay A. Vashisht,2

Albert Lee,1,6 James Wohlschlegel,2 and Guillaume F. Chanfreau1,*1Department of Chemistry and Biochemistry and the Molecular Biology Institute, University of California, Los Angeles, 607 Charles E. YoungDrive East, Box 951569, Los Angeles, CA 90095-1569, USA2Department of Biological Chemistry, University of California, Los Angeles, 310 BSRB, Box 951737, Los Angeles, CA 90095-1737, USA3Present address: EASEL Biotechnologies, 11755 Wilshire Boulevard, Los Angeles, CA 90025, USA4Present address: Genomics Institute of the Novartis Research Foundation, San Diego, CA 92121, USA5Present address: Instituto de Quımica, University de Sao Paulo, Avenida Professor Lineu Prestes 748 Butanta, 05508-900 Sao Paulo, Brazil6Present address: Agilent Technologies, 1834 State Highway 71 West, Cedar Creek, TX 78612-3412, USA*Correspondence: guillom@chem.ucla.eduhttp://dx.doi.org/10.1016/j.molcel.2013.05.005

SUMMARY

Zinc is an essential cofactor of all major eukaryoticRNApolymerases. How the activity of these enzymesis coordinated or regulated according to cellular zinclevels is largely unknown. Here we show that the sta-bility of RNA polymerase I (RNAPI) is tightly coupledto zinc availability in vivo. In zinc deficiency, RNAPI isspecifically degraded by proteolysis in the vacuolein a pathway dependent on the exportin Xpo1pand deubiquitination of the RNAPI large subunitRpa190p by Ubp2p and Ubp4p. RNAPII is unaf-fected, which allows for the expression of genesrequired in zinc deficiency. RNAPI export to thevacuole is required for survival during zinc starvation,suggesting that degradation of zinc-binding subunitsmight provide a last resort zinc reservoir. Theseresults reveal a hierarchy of cellular transcriptionalactivities during zinc starvation and show that degra-dation of the most active cellular transcriptionalmachinery couples cellular growth and proliferationto zinc availability.

INTRODUCTION

Ribosome biogenesis is an energetically expensive process(Warner, 1999) that must be tightly regulated according tocellular demands, nutrient abundance, and stress conditions.One of the most frequent strategies to regulate ribosome bio-genesis is to control the transcription of ribosomal DNA (rDNA)by RNA polymerase I (RNAPI). Negative regulation of RNAPIactivity is frequently observed when cells are exposed tonutrient deprivation or suboptimal growth conditions (Grummt,2003; Lempiainen and Shore, 2009; Warner, 1999). In general,RNAPI-mediated transcription is controlled by two major mech-anisms: epigenetic control of rDNA chromatin can modulate thenumber of active rDNA transcription units, whereas regulationof RNAPI transcription factors directly controls RNAPI activity,

typically by limiting the number of initiation-competent RNAPIcomplexes (Grummt, 2003; Moss, 2004). The latter strategy ismost frequently utilized to quickly respond to environmentalcues, for example by modulating the association of RNAPI withone of its key transcription factors such as Rrn3p, the yeastortholog of mammalian TIF-IA (Grummt, 2003; Moss, 2004).Such a mechanism is used in nutrient starvation conditions,during which the target of rapamycin (TOR) signal transductionpathway regulates RNAPI activity by reducing the number ofRNAPI complexes associated with Rrn3p or its ortholog TIF-IA(Claypool et al., 2004; Mayer and Grummt, 2006; Powers andWalter, 1999; Philippi et al., 2010). RNAPI activity can also beregulated at the elongation stage (Stefanovsky et al., 2006)and by stresses that affect the integrity of cellular components.For example, induction of DNA damage activates the ataxiatelangiectasia mutated (ATM) signal transduction pathway inmammalian cells and inhibits RNAPI activity (Kruhlak et al.,2007). Exposure of mammalian cells to oxidative stress triggersthe export of transcription factor TIF-IA to the cytoplasm(Mayer et al., 2005), resulting in RNAPI downregulation througha reduction of the number of RNAPI core complexes associatedwith TIF-IA in the nucleus. The fact that RNAPI activity isregulated during the cell cycle and in conditions of cellulardamage underscores the importance of coupling RNAPI activityto cellular activity and integrity. Recently, the stability of the largesubunit of RNAPI was shown to be modulated by ubiquitination(Richardson et al., 2012). Deubiquitination of Rpa190p bythe Ubp10p ubiquitin protease was shown to stabilize RNAPIand to enhance ribosomal RNA (rRNA) synthesis (Richardsonet al., 2012). This study showed that rRNA production can becontrolled not only by modulating the level of transcriptionallycompetent RNAPI complexes, but also by changing the stabilityof the large subunit of RNAPI.Zinc is an essential component of all three major eukaryotic

RNA polymerases, including RNAPI (Cramer et al., 2001;Donaldson and Friesen, 2000; Naryshkina et al., 2003; Treichet al., 1991), and is also found in a large number of cellularproteins. Despite a thorough characterization of the responseof eukaryotic cells to zinc deficiency (North et al., 2012), it isunclear how the activity of eukaryotic RNA polymerases isinfluenced by zinc levels, which is critical for the activity of these

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 105

enzymes. In this study, we report the unexpected finding thatRNAPI complexes are downregulated during zinc deficiencyvia export to the vacuole and proteolytic degradation, while thelevels of most RNA Polymerase II and III subunits remain un-changed. Given the importance of RNAPI activity in controllingcellular growth and proliferation (Drygin et al., 2010), degradationof RNAPI provides a unique mechanism to limit cellular growthcontrol in conditions of zinc scarcity.

RESULTS

Decrease of rRNA Synthesis during Zinc Deficiency IsDue to RNA Polymerase I DownregulationIn previous gene expression analysis of yeast cells shifted in amedium containing the nonpermeable metal chelator bathophe-nanthroline sulfonate (BPS; Lee et al., 2005), we observed adecrease in the amount of rRNAs (Figure 1A, bottom). Northernblot analysis of pre-rRNAs showed a progressive decrease inthe amount of 35S, 27S, and 7S pre-rRNA intermediates in theseconditions, suggesting a defect in transcription (Figure 1A, top).Pulse-chase analysis revealed that the transcription of rRNAswas severely reduced (Figure 1B) to levels comparable tothose of a strain deleted for ribonuclease (RNase) III, which isdeficient in rRNA transcription (Catala et al., 2008). To investigatethe mechanism of this downregulation, we monitored the levelsof RNA polymerase I subunits using C-terminally tandem affinitypurification (TAP)-tagged strains (Figure 2A). Western blot anal-ysis performed from crude extracts prepared from these taggedstrains showed that all RNAPI subunits tested were downregu-lated in cells grown with BPS. This effect was also detected forRpa135p expressed from a plasmid-borne HA-tagged versionor detected using an anti-Rpa135 monoclonal antibody (Fig-ure 2A). We observed a downregulation of the Rpb6p subunitcommon to all three RNA polymerases (Figure 2A), but mostRNAPII-specific subunits did not exhibit any change (Figure 2A).The only exception was the large subunit Rpb1p, which ap-

peared downregulated based on immunoblot analysis usingthe monoclonal antibody 8WG16 that recognizes the unphos-phorylatedC-terminal domain (CTD).Western blot analysis usingthe N200 antibody that recognizes the N terminus of Rpb1p(Chen and Hahn, 2004) showed that a shorter fragment ofRpb1p accumulates during BPS treatment (Figure 2A), suggest-ing that the CTD is cleaved. In agreement with this hypothesis,the 8WG16 antibody detected a small fragment in high-percent-age polyacrylamide gels upon BPS treatment (Figure 2C).Additional experiments showed that the apparent Rpb1p CTDcleavage is the result of a proteolytic artifact occurring duringpreparation of crude extracts (Figure S1 available online).BPS is generally used to chelate iron but can also chelate zinc

(Zhao and Eide, 1996). We found that growing yeast cells in thepresence of another divalent cation chelator, EDTA also resultedin RNAPI downregulation (Figure 2B and see below). By contrast,subunits of the two other major RNA polymerases, such as Rpb9and Rpb11 (RNAPII) and Rpc31p, Rpc34p, and Rpc82p(RNAPIII), were unaffected by growth in EDTA-containing me-dium (Figure 2B). Thus, the downregulation observed in cellsgrown in the presence of metal chelators, such as BPS orEDTA, is specific to RNAPI. These results raised the questionof which metal deficiency was responsible for the effectsobserved. To answer this question, we prepared media contain-ing low-zinc medium (LZM) or low-iron medium (LIM) by preincu-bating synthetic medium with chelating beads and adding backall metals except iron or zinc. Cells shifted into LZM, but not LIM,exhibited downregulation of the RNAPI subunits Rpa190-TAPand Rpa135p (Figure 2D), indicating that the downregulationobserved is specific to zinc deficiency. In agreement with azinc-specific effect, we found that cells preloaded with excesszinc by preculture in amedium supplemented with 2mM zinc ex-hibited a slower RNAPI downregulation than cells precultured innonsupplemented medium (Figure 2E). In addition, a fasterdownregulation of RNAPI was observed in a strain lacking theZap1p transcription factor (Figure 2F), which is genetically

Figure 1. Downregulation of Ribosomal RNASynthesis in S. cerevisiae Cells Grown in thePresence of the Metal Chelator BPS(A) Northern analysis and ethidium bromide staining

of rRNAs. Top: the membrane was probed with an

oligonucleotide hybridizing to the 35S pre-rRNA

precursor. Cross-hybridization of the oligonucleo-

tide is responsible for the faint signal observed at

the positions of the 18S and 25S rRNAs. ScR1 was

used as a loading control. Bottom: ethidium bro-

mide staining. RNA profiles obtained with wild-type

cells treated with BPS were compared to that of an

rnt1D strain grown in normal medium, which is

deficient in rRNA synthesis (Catala et al., 2008).

(B) Pulse-chase analysis. Wild-type cells grown in

normal medium or treated with BPS for 10 hr were

analyzed by transcriptional pulse-chase analysis

using 3H-uracil to monitor the kinetics of rRNA

production. An rnt1D strain grown in normal me-

dium, which is deficient in rRNA synthesis, was

included for comparison. This analysis showed that

the rate of rRNA transcription was lower in wild-type

cells shifted to BPS than in rnt1D cells.

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

106 Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc.

zinc-deficient (Lyons et al., 2000). This result shows that RNAPIdownregulation does not require the Zap1p-mediated transcrip-tional response, but that genetically depriving the cells of zincaccelerates the kinetics of RNAPI downregulation.Several RNAPI subunits have been shown to bind zinc (Treich

et al., 1991). A general instability of nuclear zinc-binding proteinsin zinc starvation conditions could account for the downregu-lation of RNAPI in these conditions. To test this hypothesis, wemonitored the levels of two nuclear zinc-binding splicing factorsunrelated to RNAPI, Sad1p and Luc7p. In contrast to what wasobserved for RNAPI, levels of Sad1p and Luc7p were unaffectedby zinc starvation (Figure 2G). Similarly, the Rpb2p, Rpb3p, andRpb9p zinc-binding subunits of RNAPII were stable (Figures 2Aand 2B). Taken together, these results demonstrate that zincdeficiency results in a specific downregulation of RNAPI.

Low-Zinc-DependentRNAPolymerase I DownregulationIs Unrelated to Stress Pathways Known to Affectits Activity and Is Not Due to Cellular DeathTo investigate whether the downregulation of RNAPI during zincdeficiency is mechanistically linked to stress conditions previ-

Figure 2. Downregulation of RNA Polymer-ase I in Zinc Deficiency(A) Western blot analysis of RNA polymerase

subunits from crude protein extracts prepared

from strains grown in the presence of BPS. PAP

antibody was used for TAP-tagged proteins. The

8WG16mAb (CTD) or N200 antibodies (NTD) were

used to detect Rpb1p, and a monoclonal anti-

Rpa135p antibody was used to detect Rpa135p.

(B) Western blot analysis of RNA polymerase

subunits during EDTA treatment. Legends as in (A).

(C) The CTD of Rpb1 is cleaved in extract from

cells grown with BPS. A western blot analysis

of Rpb1 using 8WG16 and a high-percentage

acrylamide gel is shown (lower panel).

(D) Downregulation of RNAPI subunits is due to

zinc limitation. Western blots of Rpa135p or

Rpa190-TAP levels in LZM or LIM are shown.

(E) RNAPI downregulation is slower in cells

preloaded with zinc. An Rpa135-GFP western

analysis of wild-type cells pregrown in minimal

medium with (2 mM) or without (0 mM) zinc sup-

plement and shifted in a medium containing EDTA

is shown.

(F) Rpa135p downregulation occurs faster in a

strain genetically zinc deficient. An Rpa135-GFP

western analysis in wild-type and zap1D strains is

shown.

(G) Western blot analysis of the zinc-binding

proteins Sad1p and Luc7p fused to GFP during

zinc starvation.

ously known to affect its activity (Grummt,2003; Warner, 1999), we monitoredRNAPI levels in a variety of conditions ormutants known to affect RNAPI activity.RNAPI levels were unaffected by aminoacid starvation (Figure S2A) even thoughthis condition is known to result in a

reduction in rDNA transcription (Lempiainen and Shore, 2009).This result is consistent with recent data showing thatRpa190p levels are unaffected by nutrient starvation (Richard-son et al., 2012). In addition, we did not detect any RNAPI down-regulation in cells grown in the absence of glucose (see below),which completely blocks cell division. RNAPI activity is alsoknown to be affected by defects in secretion in a pathwaydependent on the Wsc family of plasma membrane sensorsand the Pkc1p protein kinase (Li et al., 2000; Nierras andWarner,1999). To investigate if the downregulation of RNAPI observed inzinc deficiency is mechanistically connected to this response,we studied the kinetics of downregulation of Rpa135p in apkc1D strain. Although this strain exhibits lower levels of RNAPIin normal zinc conditions, zinc starvation resulted in normalRNAPI downregulation kinetics (Figure S2B), showing thatPkc1p is not involved in the zinc-dependent downregulation ofRNAPI. Similarly, RNAPI downregulation was not inhibited inwscmutants (Figure S2C), indicating that RNAPI downregulationduring zinc deficiency is unrelated to the response that occurs asa result of defects in plasma membrane synthesis or secretorypathways (Li et al., 2000; Nierras and Warner, 1999).

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 107

Previous studies had shown that the downregulation of RNAPItranscriptional activity during nutrient deprivation is mediatedby the TOR signal transduction pathway (Claypool et al., 2004;Powers and Walter, 1999). To investigate if the downregulationof RNAPI during zinc deficiency is mechanistically dependenton the TOR pathway, we used tor1D or fpr1D strains deficientin TOR signaling. We found that RNAPI downregulation duringzinc deficiency does not require an intact TOR pathway, as itoccurs normally in tor1D or fpr1D mutants (Figure S2D). Takentogether, these results show that RNAPI downregulation duringzinc deficiency is unrelated to regulatory pathways previouslydescribed to affect ribosomebiogenesis or integrity. Additionally,we found that RNAPI downregulation in zinc deficiency is not

Figure 3. Stabilization of RNAPI Subunitsand Associated Proteins during Zinc Defi-ciency in Vacuolar Protease Mutants(A) Northern analysis of RPA135, RPA49, RPA43,

and ZRT1 mRNAs during zinc deficiency. SCR1

was used as a loading control.

(B) Immunoblot analysis of Rpa135p in wild-type

and vacuolar protease mutants.

(C) Same as (B), but Rpa135-GFP and Rpa43-GFP

were used.

(D) TAP purification of RNAPI complexes in an

Rpa190p-TAP pep4D strain before or after a shift in

low-zinc conditions. Extracts prepared from strains

grown in normal medium or after 8 hr in an EDTA-

containing medium were used for TAP purification.

The protein eluates after TAP purification visualized

by SYPRO Ruby staining are shown.

(E) Analysis of RNAPI-associated proteins and

other nuclear proteins during a low-zinc shift. Cells

expressing TAP-tagged subunits of proteins

known to physically associate with RNAPI (Reb1p,

Fpr3p, Rrp5p, Cbf5p; Fath et al., 2000), and other

nucleolar (Fob1p) or nuclear (Tfg1p, Smd1p,

Smb1p) proteins, were shifted in a medium con-

taining EDTA for the indicated times, and protein

levels were analyzed. Extracts prepared from

wild-type cells were also analyzed by immuno-

blotting using anti-Nop1p monoclonal antibodies.

RNAPI-associated proteins (Reb1p, Fpr3p, Rrp5p,

Cbf5p) showed downregulation, while other con-

trol proteins (Tfg1p, Smd1p, Smb1p, Fob1p,

Nop1p) were stable.

(F) Dependency on Pep4p and Prb1p for the

downregulation observed for Reb1p, Fpr3p,

Rrp5p, and Cbf5p. Legends as in (A).

due to cell death following prolongedexposure to low-zinc conditions. Whencells incubated in zinc-deficient mediumwere shifted back to normal medium,they quickly resumed growth and recov-ered RNAPI levels (Figure S3).

RNAPI Is Exported to the Vacuoleand Degraded by VacuolarProteases in Zinc DeficiencyThe downregulation of RNA polymerase Isubunits could be due to transcriptional

repression of the genes encoding these subunits or to post-transcriptional processes. We monitored the messenger RNA(mRNA) levels of genes encoding three RNAPI subunits(RPA135, RPA49, and RPA43) and found that their levelswere unaffected during a shift to LZM (Figure 3A). During thetime course of this experiment, the ZRT1 mRNA was robustlyinduced (Figure 3A). Given the short half-life of this mRNA(Toesca et al., 2011), the steady accumulation of this mRNAcombined with the observation that RNAPI mRNAs are stablyexpressed shows that that RNAPI downregulation is not an in-direct consequence of a general decrease in RNAPII-mediatedtranscription in zinc deficiency and is not due to transcriptionalrepression of RNAPI subunit genes or a degradation of RNAPI

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

108 Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc.

subunit mRNAs. We next hypothesized that this downregu-lation was due to increased protein turnover and searched forproteases involved in zinc deficiency. Vacuolar proteaseswere previously shown to be upregulated during zinc deficiency(Lyons et al., 2000). To test their involvement in RNAPI downre-gulation, we monitored Rpa135p levels in the vacuolar proteasemutant strains prb1D, pep4D, or prc1D during a shift to LZM.Figure 3B shows that Rpa135p downregulation in low zincwas rescued by inactivating Prb1p or Pep4p, but not Prc1p.The observation that inactivation of either Pep4p or Prb1pwas sufficient to rescue the downregulation of RNAPI can beexplained by the mutual requirement of these proteases foreach other for proteolytic processing to their fully functionalmature forms (Hirsch et al., 1992; Moehle et al., 1989). Simi-larly, the downregulation of GFP-tagged versions of Rpa135por Rpa43p was rescued in a prb1D strain (Figure 3C). Thiseffect was not due to a general involvement of vacuolar pro-teases in RNAPI subunit turnover under normal conditions,

Figure 4. Export of RNA Polymerase I to theVacuole during Zinc Deficiency Is Depen-dent on Xpo1p(A) Localization of Rpa49-GFP in wild-type,

pep4D, or prb1D strains in normal medium (0 hr in

EDTA) or after 8 hr in EDTA. Green, Rpa49-GFP

localization; red, vacuolar membrane stained with

FM4-64.

(B) Localization of Nop1-GFP in normal and low-

zinc conditions. Legends as in (A), except the

strains were transformed with a plasmid ex-

pressing Nop1-GFP.

(C) Genetic inactivation of Xpo1p rescues RNAPI

downregulation. Wild-type or xpo1-1 cells were

shifted for 2 hr at 37!C before a shift to EDTA, and

the level of Rpa135-GFP was analyzed by western

blot.

(D) Analysis of Rpa135-GFP localization in wild-

type and xpo1-1 strains grown at 25!C and shifted

to 37!C for 2 hr or to 37!C with EDTA.

since RNAPI subunit half-life wasidentical in wild-type and vacuolar prote-ase mutants grown in normal medium(Figure S4). Based on these results, wehypothesized that RNAPI complexesare exported to the vacuole anddegraded by vacuolar proteases duringzinc deficiency. Using RNAPI subunitsfused to GFP to monitor RNAPI localiza-tion during zinc starvation, we observeda major loss of GFP signal in wild-typecells (Figure 4A), consistent with theimmunoblot analysis. By contrast, GFP-fused RNAPI accumulated inside thevacuolar membrane (stained with thered vital stain FM4-64; Vida and Emr,1995) in the pep4D or prb1D strains(Figure 4A). This change of localizationis highly specific to RNAPI and was not

observed with the nucleolar protein Nop1p (Figure 4B), whichis present in and physically associates with the same nucleolarfraction as RNAPI (Fath et al., 2000). These experimentsdemonstrate that RNAPI is specifically exported to the vacuoleduring zinc deficiency, where it is degraded by vacuolar prote-ases. To investigate whether RNAPI disassembles prior tovacuolar export, we performed a TAP-tag purification (Rigautet al., 1999) of RNAPI complexes from extracts preparedfrom a pep4D strain expressing Rpa190p-TAP grown in normalmedium or EDTA. As shown in Figure 3D, the pattern of bandsobtained after TAP purification was very similar before and afterthe shift to zinc deficiency. In addition, by mass spectrometrywe could detect the presence of all RNAPI subunits in thesepurified complexes in sub-stoichiometric proportions beforeand after the shift (Table S1). Furthermore, several proteinsthat were shown to be present in the same nucleolar fractionand/or were coimmunoprecipitated with RNAPI, such asReb1p, Fpr3p, Rrp5p, and Cbf5p (Fath et al., 2000), were

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 109

degraded in the same Pep4p/Prb1p-dependent manner duringzinc deficiency (Figures 3E and 3F). The downregulation ofsome RNAPI-associated rRNA processing factors, such asCbf5p and Rrp5p, might contribute to the defects in rRNA syn-thesis described in Figure 1. By contrast, several other nuclearand nucleolar proteins were unaffected (Tfg1p, Smd1p,Smb1p, Fob1p), including Nop1p (Figure 3E). Interestingly,although Nop1p is coimmunoprecipitated by RNAPI (Fathet al., 2000), it remains localized in the nucleolus during zincstarvation (Figure 4B) and is not downregulated (Figure 3E).This result shows that only a specific subset of RNAPI-associ-ated proteins is degraded with RNAPI. We conclude thatRNAPI and some of its associated proteins are likely to be ex-ported during zinc deficiency as a complex to the vacuole,where they are degraded by vacuolar proteases.

Export of RNAPI to the Vacuole Requiresthe Exportin Xpo1p but Is Independentof Known Autophagy PathwaysWe next investigated whether RNAPI complexes are exportedfrom the nucleus to the vacuole via nucleus-vacuole junctions,which are formed during piecemeal microautophagy of thenucleus (PMN) in starvation conditions (Kvam and Goldfarb,2007). The nvj1D and vac8D mutant strains defective in thePMN pathway did not rescue RNAPI downregulation duringzinc starvation (Figure S2E), showing that PMN is not requiredfor translocation of RNAPI to the vacuole. We next investigateda potential function of the nuclear export factor Xpo1p(Stade et al., 1997) in this process and found that geneticinactivation of Xpo1p, prior to a shift to EDTA, inhibited thedownregulation of RNAPI (Figure 4C) by preventing RNAPIexport out of the nucleus (Figure 4D). Although we cannotcompletely rule out that RNAPI vacuolar degradation is due toa failure of newly synthesized protein to fold or assemble duringzinc starvation, these results support the hypothesis that Xpo1pmediates the export of the RNAPI complex out of the nucleus forsubsequent import into the vacuole during zinc deficiency. Tounderstand the mechanism of vacuolar import of RNAPI duringzinc deficiency, we screened mutant strains defective in variousvacuolar import pathways. Neither of the atg mutants analyzed(atg1D, atg7D), which are defective in the autophagy or cyto-plasm-to-vacuole targeting (CVT) pathways, showed a rescueor delay in the downregulation of RNAPI subunits during zincdeficiency (Figure S2F). Thus, RNAPI vacuolar degradation inzinc deficiency does not require any of the known autophagyor CVT pathways.

Downregulation and Vacuolar Import of RNAPIin Zinc Deficiency Requires Deubiquitinationof the Large Subunit Rpa190pRecent results showed that stability of the RNAPI large subunitRpa190p is affected by ubiquitination (Richardson et al., 2012).To investigate if this modification is involved in RNAPI vacuolardegradation during zinc deficiency, we probed blots of TAP-purified RNAPI complexes (prepared from a pep4D strain)with an anti-ubiquitin antibody. We detected high molecularmass (>190 kDa) bands (Figure 5A), which, given their mobility,were likely to correspond to ubiquitinated Rpa190p. These

bands disappeared after 8 hr in LZM, suggesting thatRpa190p becomes deubiquitinated in low-zinc conditions. Tofurther demonstrate this, we used a strain expressing His-tagged ubiquitin and an HA-tagged version of Rpa190p (Ri-chardson et al., 2012). After introducing a pep4D deletion inthis strain, we purified ubiquitinated proteins under denaturingconditions using Ni-agarose beads. We confirmed the pres-ence of ubiquitinated HA-tagged Rpa190p in normal growthconditions (Figure 5B). However, we observed that the levelsof ubiquitinated Rpa190p decreased during zinc starvation (Fig-ure 5B). This decrease could also be detected in crude extractsbased on the disappearance of the high-molecular-mass smearin low-zinc conditions (Figure 5B). These observations suggestthat deubiquitination of Rpa190p might be required for nuclearexport and subsequent vacuolar degradation of the RNAPIcomplexes. To test this hypothesis, we analyzed a number ofubiquitin protease mutants to determine their ability to mediateRNAPI degradation in low zinc. We found that inactivation ofthe Ubp2p and Ubp4p deubiquitinases, which are requiredfor sorting proteins to the vacuole (Amerik et al., 2000; Dupreand Haguenauer-Tsapis, 2001), resulted in greater stability ofRNAPI in low-zinc conditions (Figure 5C). By contrast, inactiva-tion of Ubp3p had no effect (Figure 5C), showing that RNAPIvacuolar degradation does not require Ubp3p, and thus ismechanistically unrelated to ribophagy, a process in whichmature ribosomes are targeted to the vacuole during starvation(Kraft et al., 2008). Similarly, Rpa190p levels were not rescuedupon inactivation of Ubp8p (Figure 5C) and Ubp14p (Fig-ure S5A), showing that the stabilization observed in theubp2D and ubp4D strains was specific to these two ubiquitinproteases. Recently, Ubp10p was shown to stabilizeRpa190p (Richardson et al., 2012), suggesting that this ubiqui-tin protease might also be involved in controlling RNAPI stabil-ity during zinc starvation. However, we found that Ubp10pinactivation had no effect on Rpa190p downregulation duringzinc starvation (Figure 5C). This result demonstrates that thedeubiquitination events involved in stabilizing RNAPI duringnormal growth conditions and promoting degradation in zincstarvation are mechanistically distinct. The stabilization ofRNAPI in the ubp2D and ubp4D strains was specific to zincstarvation, as inactivation of these ubiquitin proteases had noeffect on Rpa190p levels during glucose or amino acid starva-tion (Figure S5B). Finally, we found that growth of ubp2D andubp4D strains in low zinc resulted in inhibition of RNAPI vacu-olar import, as the majority of GFP-tagged Rpa190p localizedoutside the vacuole in these strains (Figure 6A). These experi-ments demonstrate that the activity of the Ubp2p and Ubp4pubiquitin proteases is required for vacuolar import and degra-dation of RNAPI during zinc deficiency and is consistent withtheir demonstrated role in sorting vacuolar proteins and multi-vesicular bodies (Amerik et al., 2000; Dupre and Haguenauer-Tsapis, 2001).

Inactivation of the Ubiquitin Ligase Rsp5p UncouplesDegradation of RNAPI from Zinc DeficiencyThe Rsp5p ubiquitin ligase was previously implicated in rRNAbiogenesis and ribosome stability (Neumann et al., 2003;Shcherbik and Pestov, 2011). In addition, Rsp5p exhibits a

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

110 Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc.

nuclear localization and was found to interact with RNA poly-merase subunits (Neumann et al., 2003). Thus, Rsp5p wasa prime candidate to mediate the ubiquitination of RNAPI.If deubiquitination of RNAPI is a major determinant of RNAPIdegradation in the vacuole, we predicted that inactivation ofRsp5p might result in Pep4p/Prb1p-dependent degradation ofRNAPI complexes independently of zinc levels. Indeed, theabundance of Rpa190p quickly decreased in a strain carryingan rsp5-1 temperature-sensitive (ts) allele grown in normalmedium but shifted to nonpermissive temperature (Figure 5D).

Figure 5. Loss of Rpa190p UbiquitinationTriggers RNAPI Downregulation duringZinc Deficiency(A) Analysis of ubiquitin in TAP-purified RNAPI

complexes. RNAPI was TAP purified from

pep4D cells expressing a TAP-tagged version of

Rpa190p grown in normal medium or after 4 or

8 hr in EDTA. TAP-purified RNAPI complexes were

visualized by SYPRO Ruby staining (right)

or analyzed by immunoblot using anti-ubiquitin

antibody or anti-Rpa135 monoclonal antibody.

Whole-cell extracts were analyzed by immunoblot

using the anti-ubiquitin antibody.

(B) Rpa190p is deubiquitinated in low-zinc condi-

tions. Immunoblots of ubiquitinated proteins

purified by chromatography on Ni-agarose beads

from extracts prepared from strains expressing

His-tagged ubiquitin and HA-tagged Rpa190p

are shown. Two exposures of the same blot are

shown.

(C) Immunoblot analysis of RNAPI downregulation

in low zinc in various ubiquitin proteases mutant

strains.

(D) Immunoblot analysis of RNAPI levels in the

rsp5-1 and rsp5-1pep4D mutants at permissive

and nonpermissive temperatures.

Strikingly, this downregulation wasabolished when a PEP4 deletion wasintroduced in this mutant (Figure 5D).We monitored Rpa190p localization inthese strains grown in normal medium,before or after a shift to 37!C. Unlikethe wild-type control, we observed aloss of GFP signal in the rsp5-1 mutantshifted to 37!C (Figure 6B). However,in the rsp5-1 pep4D double mutant, theGFP signal was recovered and accu-mulated in both the vacuole and thenucleolus (Figure 6B). The partial nucle-olar signal is likely due to incompleteinactivation of Rsp5p at restrictive tem-perature. Taken together, these resultsshow that genetic inactivation of Rsp5presults in vacuolar degradation of RNAPIthat is mechanistically indistinguishableto what is observed during zinc defi-ciency and demonstrate that deubiquiti-nation of Rpa190p is a main determinant

in triggering vacuolar export and degradation of the RNAPIcomplex.

The Ubp2p and Ubp4p Ubiquitin Proteases Are Requiredfor Optimal Fitness or Survival in Zinc StarvationConditionsRNA polymerase I contains multiple zinc-binding subunits(Treich et al., 1991). Because the vacuole is the site of zinc stor-age and redistribution (Simm et al., 2007), we hypothesized thatexport of RNAPI to the vacuole for degradation might provide a

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 111

last-resort zinc reservoir during zinc starvation. If this hypothesiswas true, inhibiting export of RNAPI was predicted to severelyhamper cellular growth in zinc starvation conditions by limitingthe amount of zinc that could be recovered from RNAPI zinc-binding subunits. Indeed, the ubp2D strain exhibited a growthdefect, which could be remedied by zinc supplementation, in amedium containing EDTA (Figure 7). More strikingly, the ubp4Dstrain, which showed a severe growth defect in normal media(synthetic medium [SDC] or rich medium [YPD]), was inviablein the presence of EDTA, but its growth could be restored byadding back zinc. These growth phenotypes were specific tozinc starvation, as the ubp2D and ubp4D strains did not exhibitany specific growth defect in other stress conditions, such ashigh iron or copper (Figure 7), and grew similarly to a wild-typestrain in amino acid or glucose starvation conditions (FigureS5B).In contrast, the pep4D strain did not show any growth defectin the presence of EDTA (Figure 7), suggesting that degradationof RNAPI might not be required to release zinc from RNAPIsubunits once they have been imported to the vacuole. Takentogether, these data demonstrate that the export of RNAPIto the vacuole mediated by Ubp2p and Ubp4p is required forefficient growth in the absence of zinc.

DISCUSSION

The results presented in this study show that zinc deficiencyresults in vacuolar degradation of RNAPI and some of itsassociated proteins in the yeast S. cerevisiae. This regulationdramatically decreases rRNA transcription, which quantitativelyaccounts for most of the transcriptional activity in rapidlygrowing cells. The levels of most RNAPII subunits were unaf-fected (Figure 2), and the transcription of the RNAPI subunitsand ZRT1 mRNAs were robust (Figure 3A), showing thatRNAPII transcription is fully functional in low-zinc conditions toallow for the expression of genes induced during zinc deficiency(Lyons et al., 2000). Thus, despite the apparent downregulationof the Rpb6p subunit shared by all three RNA polymerases (Fig-ure 2A), the level of this subunit must be sufficient to fulfill theRNAPII activity necessary to robustly express low zinc-responsegenes. These results are consistent with previous observationsin Euglena showing that zinc deficiency results in the formationof only one peak of RNA polymerase activity, which is likely tobe an a-amanitin-resistant form of RNAPII (Falchuk et al.,1985). Our results might also shed light on older studies showingthat feeding rats a zinc-deficient diet results in overall reducedRNA polymerase activity (Terhune and Sandstead, 1972);however, these results could be due to zinc depletion from thepolymerases rather than the active degradation mechanismdemonstrated here. Similar responses to lower RNAPI activityin zinc deficiency across different eukaryotic lineages suggestthe existence of a conserved strategy of zinc economy thatslows cellular growth and spares an essential metal for proteinswhose functions are needed to survive these adverse conditions.The proteins that we identified in this study as involved in RNAPIdownregulation were not found in a large-scale study that char-acterized mutant strains in zinc deficiency (North et al., 2012).This discrepancy might be explained by the fact that, in theirstudy, North et al. investigated the effect of limiting zinc con-centrations, whereas in this study, zinc is completely lacking.Thus, the factors that influence survival in each of these growthconditions might be different.

A Mechanism for RNAPI Degradation in Zinc DeficiencyOur results support a model in which Rpa190p deubiquitinationtriggers vacuolar import of the RNAPI complex. This regulatedvacuolar degradation can be uncoupled from zinc deficiencyby inactivating Rsp5p (Figure 5D). The destabilization of theRNAPI complex in the rsp5-1 mutant might explain some ofthe phenotypes described in this mutant, such as a decreasein rRNA biogenesis and a reduction in overall ribosome levels(Shcherbik and Pestov, 2011). This result contrasts with the gen-eral role of ubiquitination in triggering protein degradation as wellas with the recent finding that deubiquitination of Rpa190p byUbp10p strongly stabilizes Rpa190p (Richardson et al., 2012).The likely explanation is that the mono- or polyubiquitinationsites that destabilize RNAPI under normal growth conditions(Richardson et al., 2012) are distinct from the sites that stabilizeRNAPI and are deubiquitinated under low-zinc conditions (thisstudy). This hypothesis is supported by the observation thatUpb10p has no influence on RNAPI downregulation during zincstarvation (Figure 5C). One specific mechanism that remains to

Figure 6. Analysis of RNA Polymerase I Localization in ubp2D,ubp4D, and rsp5-1 Mutants(A) Analysis of Rpa190-GFP localization in wild-type, ubp2D, and ubp4D

mutants. Rpa190-GFP (green) localization was compared to the vacuole (red)

stained with the vital dye FM4-64 before (T = 0) or after (T = 8) a shift to low-zinc

conditions.

(B) Analysis of Rpa135-GFP localization in wild-type, rsp5-1, and rsp5-1pep4D

grown at 25!C or shifted to 37!C for 5 hr.

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

112 Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc.

be determined is what controls the deubiquitination of RNAPIunder low-zinc conditions. A decrease in intracellular zinc con-centration might trigger a sensing mechanism that exposes theRNAPI complex to the action of ubiquitin proteases for vacuolarimport. Such a mechanism might involve conformationalchanges in the RNAPI complex upon a loss of zinc by somezinc-binding RNAPI subunits, which might trigger the bindingof Ubp2p or Ubp4p. Alternatively, it is possible that the activityof one of the proteins that promote Rsp5p-mediated ubiquitina-tion or Ubp2p/Ubp4p-mediated deubiquitination of RNAPI isdirectly controlled by zinc level, which would directly affect deu-biquitination of Rpa190p.

Why Degrade RNAPI in Zinc Deficiency?A frequent strategy adopted by cells duringmetal deficiency is toreduce the expression or stability of specific metalloproteins inorder to spare particular metals and/or redistribute them to otherproteins (Bird et al., 2006; Merchant and Bogorad, 1986; Puiget al., 2005). Biochemical and structural studies have shownthat many subunits of all three eukaryotic RNA polymerases,including RNAPI, contain multiple zinc atoms (Cramer et al.,2001; Donaldson and Friesen, 2000; Naryshkina et al., 2003;Treich et al., 1991). Given the large amounts of RNAPI subunitsin the cell ("104/cell; Ghaemmaghami et al., 2003), degradationof these subunits in the vacuole, the site of zinc storage andredistribution (Simm et al., 2007), might provide a last-resortzinc reservoir during prolonged zinc starvation. This model issupported by the observation that the ubp2D and ubp4D strainsexhibit specific growth defects during zinc starvation (Figure 7).The degradation of RNA polymerase subunits during zinc defi-ciency might represent an extreme example of metal sparing,indicating that the functions of other cellular zinc-bindingproteins are prioritized over those involved in ribosomal RNAproduction. In addition, downregulation of RNAPI and rRNA

Figure 7. Growth of Wild-Type and Ubiqui-tin ProteaseMutant Strains in NormalMediaand Various Stress ConditionsStrains shown were derived from a strain ex-

pressing His-tagged ubiquitin and an HA-tagged

Rpa190p, in which the PEP4, UBP2, or UBP4

genes were knocked out by targeted disruption.

The plates were incubated on the different media

for the indicated days at 30!C.

transcription would ultimately result inribosome depletion and turnover, whichcould also provide a zinc supply giventhe abundance of zinc-binding proteinsin ribosomes. When zinc is rate limitingfor growth, degradation of zinc-richmacromolecules, such as RNAPI andribosomes, may provide an efficientmechanism by which to coordinatelyslow cellular growth, while at the sametime release an intracellular zinc reservoirto proteins needed for cell survival in lowzinc. We speculate that some of this zinc

might be redistributed to RNAPII to allow for expression ofgenes required for survival in these conditions. Because RNAPIdownregulation is quickly reversible (Figure S3), it provides cellswith the ability to fully restore ribosome biogenesis when zinc isreintroduced to cells. The finding that yeast cells degrade themost active transcriptional machinery during zinc deficiencyunderscores the importance of zinc economy in cellular meta-bolism and demonstrates a uniquemechanism of cellular growthcontrol.

EXPERIMENTAL PROCEDURES

For zinc starvation, S. cerevisiae cells were grown to log phase in synthetic

defined (SDC) media, washed twice with sterile water, and shifted to SDmedia

containing either 50 mM BPS or 1 mM EDTA for the indicated times. For

temperature-sensitive mutants, cells were first grown in SD at 25!C and

shifted to 37!C for the indicated times. For the temperature-sensitive mutants

in low-zinc conditions, cells were grown to log phase at 25!C and shifted to

37!C for 2 hr. Cells were then washed twice with prewarmed (37!C) sterile

water and shifted to prewarmed LZM for the indicated times.

Protein and RNA AnalysisFor protein analysis, cells were harvested and resuspended in lysis buffer

(200 mM Tris-HCl [pH 8.0], 320 mM ammonium sulfate, 5 mM MgCl2,

10 mM EGTA [pH 8.0], 20 mM EDTA [pH 8.0], 1 mM dithiothreitol [DTT],

20% glycerol, 1 mM phenylmethylsulfonyl fluoride [PMSF], 2 mM benzami-

dine HCl, and protease inhibitor cocktail) and vortexed with glass beads

for 8 min at 4!C. Supernatants were collected by centrifugation, and total

protein concentration was measured using the Bradford method. Protein

samples from these crude extracts were analyzed by 8% SDS-PAGE and

transferred to polyvinylidene difluoride (PVDF) membranes for western blot

analysis. Antibodies used for western blot analysis were as follows: anti-

Rpa135 (1:5,000; gift from M. Oakes and M. Nomura), 8WG16 (1:5,000;

Covance), N200 antibody (1:2,500; gift from Steve Hahn), JL-8 anti-GFP

(1:10,000; Clontech Laboratories), peroxidase antiperoxidase (PAP;

1:1,000; Rockland), anti-HA (1:10,000; Santa Cruz), anti-Nop1 (1:10,000;

EnCor Biotechnology), and anti-ubiquitinylated proteins (1:5,000; clone

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 113

FK2; Millipore). All antibodies were diluted in blocking solution (13 PBS with

Tween 20 [PBST], 5% milk).

RNA extraction and northern analysis was performed as described previ-

ously (Lee et al., 2005). We denatured 10 mg of total RNA with either formalde-

hyde buffer (formamide and formaldehyde) or glyoxal treatment in which total

RNA was incubated with glyoxal mixture (DMSO, glyoxal, PIPES, Bis-Tris,

EDTA, and glycerol) at 55!C for 1 hr. The denatured total RNAs were then

resolved on 1.2% agarose gels. Gels were transferred to nylon membrane

(N+; GE Healthcare). Probes were generated by PCR products or by 50 end

labeling of oligonucleotides for rRNA, and hybridization was performed at

65!C overnight.

Tandem Affinity PurificationTAP purification was performed as described previously (Rigaut et al., 1999)

with the following modifications. First, cells were harvested and resuspended

in IPP50 or IPP150 buffer with 25% glycerol. Cell suspension was immediately

frozen in liquid N2 and stored at #80!C. Cells were lysed by manual grinding

with a mortar and pestle in liquid N2. Grinding and freezing was repeated until

the cells became powder-like. Cell lysates were thawed on ice before centri-

fugation. Second, calmodulin beads were incubated with elution buffer at

4!C for 30 min before samples were eluted.

Vacuolar Membrane Staining and MicroscopyCells expressing RNAPI-GFP or Nop1p-GFP fusion proteins were grown to

log phase, harvested, resuspended in fresh medium, and labeled with the

vacuolarmembrane dye FM4-64 (20 mM; Invitrogen; Vida and Emr, 1995). Cells

were incubated with shaking at 30!C for 20 min and washed with fresh media

twice to remove excess dye. Cells were then resuspended into fresh medium

at 30!C for 2 hr and analyzed by fluorescence microscopy. For low zinc or

nonpermissive temperature shift, cells were first shifted to LZM or nonpermis-

sive temperature. At 2 hr prior to the indicated times, cells were harvested

and labeled with FM4-64 dye as described above for 2 hr and visualized

immediately by fluorescent microscope.

Ubiquitin Pull-Down in Denaturing ConditionsCells expressing Rpa190-3HA and 8HIS-Ubiquitin (Richardson et al., 2012)

were grown to optical density 600 (OD600) = 0.8. A total of 250 ml of cells

were harvested and resuspended in 40 mL of cold lysis buffer (1.85 N

NaOH, 7.5% 2-mercaptoethanol) and incubated on ice for 10 min. Proteins

were precipitated by 100% trichloroacetic acid (TCA) (final 20% TCA) on ice

for 10 min. A protein pellet was obtained by centrifugation and washed with

ice-cold acetone. The pellet was resuspended in buffer A (6 M guanidine

hydrochloride, 100 mM Tris-HCl [pH 8.0], 20 mM imidazole). Lysates were

clarified by centrifugation. Cleared lysates were incubated with nickel-

nitrilotriacetic acid (Ni-NTA) agarose beads (QIAGEN) overnight at room

temperature. Proteins bound to nickel beads were washed with buffer A and

buffer B (8 M urea, 100 mM Tris-HCl [pH 8.0], 20 mM imidazole) three times

each. Bound proteins were eluted from nickel beads by heating at 100!C for

20 min in SDS loading buffer and analyzed by SDS-PAGE and western blot

analysis.

SUPPLEMENTAL INFORMATION

Supplemental Information includes five figures and one table and can be found

with this article online at http://dx.doi.org/10.1016/j.molcel.2013.05.005.

ACKNOWLEDGMENTS

We thank M. Oakes and M. Nomura for the gift of anti-Rpa135p monoclonal

antibody; R. Gardner, A. Goldfarb, S. Hahn, C. Guthrie, E. Jones, M. Lund,

K. Severinov, J. Warner, and C. Woolford for gifts of plasmids, strains, or

antibodies; M. Carey, E. de Robertis, S. Merchant, G. Payne, E. Gralla, and

J. Valentine for helpful discussions; D. Eide for sharing results prior to publica-

tion; and M. Carey, A. Courey, J. Coller, and K. Roy for comments on the

manuscript. C.Y.L. was supported by a Ruth L. Kirschstein National Research

Service Award (GM07185) and by a UCLA Dissertation Year Fellowship. A.L.

was supported by the UCLA Chemistry-Biology Interface training program

and by a Fellowship from the American Heart Association Western States

Affiliate. G.C. and J.W. were supported by NIH grants GM061518 and

GM089778, respectively.

Received: January 8, 2013

Revised: March 28, 2013

Accepted: May 2, 2013

Published: June 6, 2013

REFERENCES

Amerik, A.Y., Nowak, J., Swaminathan, S., and Hochstrasser, M. (2000). The

Doa4 deubiquitinating enzyme is functionally linked to the vacuolar protein-

sorting and endocytic pathways. Mol. Biol. Cell 11, 3365–3380.

Bird, A.J., Gordon, M., Eide, D.J., andWinge, D.R. (2006). Repression of ADH1

and ADH3 during zinc deficiency by Zap1-induced intergenic RNA transcripts.

EMBO J. 25, 5726–5734.

Catala, M., Tremblay, M., Samson, E., Conconi, A., and Abou Elela, S. (2008).

Deletion of Rnt1p alters the proportion of open versus closed rRNA gene re-

peats in yeast. Mol. Cell. Biol. 28, 619–629.

Chen, H.T., and Hahn, S. (2004). Mapping the location of TFIIB within the RNA

polymerase II transcription preinitiation complex: a model for the structure of

the PIC. Cell 119, 169–180.

Claypool, J.A., French, S.L., Johzuka, K., Eliason, K., Vu, L., Dodd, J.A., Beyer,

A.L., and Nomura, M. (2004). Tor pathway regulates Rrn3p-dependent recruit-

ment of yeast RNA polymerase I to the promoter but does not participate in

alteration of the number of active genes. Mol. Biol. Cell 15, 946–956.

Cramer, P., Bushnell, D.A., and Kornberg, R.D. (2001). Structural basis of tran-

scription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863–

1876.

Donaldson, I.M., and Friesen, J.D. (2000). Zinc stoichiometry of yeast RNA po-

lymerase II and characterization of mutations in the zinc-binding domain of the

largest subunit. J. Biol. Chem. 275, 13780–13788.

Drygin, D., Rice, W.G., and Grummt, I. (2010). The RNA polymerase I transcrip-

tion machinery: an emerging target for the treatment of cancer. Annu. Rev.

Pharmacol. Toxicol. 50, 131–156.

Dupre, S., and Haguenauer-Tsapis, R. (2001). Deubiquitination step in the

endocytic pathway of yeast plasma membrane proteins: crucial role of

Doa4p ubiquitin isopeptidase. Mol. Cell. Biol. 21, 4482–4494.

Falchuk, K.H., Mazus, B., Ber, E., Ulpino-Lobb, L., and Vallee, B.L. (1985). Zinc

deficiency and the Euglena gracilis chromatin: formation of an alpha-amanitin-

resistant RNA polymerase II. Biochemistry 24, 2576–2580.

Fath, S., Milkereit, P., Podtelejnikov, A.V., Bischler, N., Schultz, P., Bier, M.,

Mann, M., and Tschochner, H. (2000). Association of yeast RNA polymerase

I with a nucleolar substructure active in rRNA synthesis and processing.

J. Cell Biol. 149, 575–590.

Ghaemmaghami, S., Huh, W.K., Bower, K., Howson, R.W., Belle, A.,

Dephoure, N., O’Shea, E.K., and Weissman, J.S. (2003). Global analysis of

protein expression in yeast. Nature 425, 737–741.

Grummt, I. (2003). Life on a planet of its own: regulation of RNA polymerase I

transcription in the nucleolus. Genes Dev. 17, 1691–1702.

Hirsch, H.H., Schiffer, H.H., and Wolf, D.H. (1992). Biogenesis of the yeast

vacuole (lysosome). Proteinase yscB contributes molecularly and kinetically

to vacuolar hydrolase-precursor maturation. Eur. J. Biochem. 207, 867–876.

Kraft, C., Deplazes, A., Sohrmann, M., and Peter, M. (2008). Mature ribosomes

are selectively degraded upon starvation by an autophagy pathway requiring

the Ubp3p/Bre5p ubiquitin protease. Nat. Cell Biol. 10, 602–610.

Kruhlak, M., Crouch, E.E., Orlov, M., Montano, C., Gorski, S.A., Nussenzweig,

A., Misteli, T., Phair, R.D., and Casellas, R. (2007). The ATM repair pathway

inhibits RNA polymerase I transcription in response to chromosome breaks.

Nature 447, 730–734.

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

114 Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc.

Kvam, E., andGoldfarb, D.S. (2007). Nucleus-vacuole junctions and piecemeal

microautophagy of the nucleus in S. cerevisiae. Autophagy 3, 85–92.

Lee, A., Henras, A.K., and Chanfreau, G. (2005). Multiple RNA surveillance

pathways limit aberrant expression of iron uptake mRNAs and prevent iron

toxicity in S. cerevisiae. Mol. Cell 19, 39–51.

Lempiainen, H., and Shore, D. (2009). Growth control and ribosome biogen-

esis. Curr. Opin. Cell Biol. 21, 855–863.

Li, Y., Moir, R.D., Sethy-Coraci, I.K., Warner, J.R., and Willis, I.M. (2000).

Repression of ribosome and tRNA synthesis in secretion-defective cells is

signaled by a novel branch of the cell integrity pathway. Mol. Cell. Biol. 20,

3843–3851.

Lyons, T.J., Gasch, A.P., Gaither, L.A., Botstein, D., Brown, P.O., and Eide,

D.J. (2000). Genome-wide characterization of the Zap1p zinc-responsive

regulon in yeast. Proc. Natl. Acad. Sci. USA 97, 7957–7962.

Mayer, C., and Grummt, I. (2006). Ribosome biogenesis and cell growth:

mTOR coordinates transcription by all three classes of nuclear RNA polymer-

ases. Oncogene 25, 6384–6391.

Mayer, C., Bierhoff, H., and Grummt, I. (2005). The nucleolus as a stress

sensor: JNK2 inactivates the transcription factor TIF-IA and down-regulates

rRNA synthesis. Genes Dev. 19, 933–941.

Merchant, S., andBogorad, L. (1986). Rapid degradation of apoplastocyanin in

Cu(II)-deficient cells of Chlamydomonas reinhardtii. J. Biol. Chem. 261, 15850–

15853.

Moehle, C.M., Dixon, C.K., and Jones, E.W. (1989). Processing pathway for

protease B of Saccharomyces cerevisiae. J. Cell Biol. 108, 309–325.

Moss, T. (2004). At the crossroads of growth control; making ribosomal RNA.

Curr. Opin. Genet. Dev. 14, 210–217.

Naryshkina, T., Bruning, A., Gadal, O., and Severinov, K. (2003). Role of

second-largest RNA polymerase I subunit Zn-binding domain in enzyme

assembly. Eukaryot. Cell 2, 1046–1052.

Neumann, S., Petfalski, E., Brugger, B., Grosshans, H., Wieland, F., Tollervey,

D., and Hurt, E. (2003). Formation and nuclear export of tRNA, rRNA and

mRNA is regulated by the ubiquitin ligase Rsp5p. EMBO Rep. 4, 1156–1162.

Nierras, C.R., andWarner, J.R. (1999). Protein kinase C enables the regulatory

circuit that connects membrane synthesis to ribosome synthesis in

Saccharomyces cerevisiae. J. Biol. Chem. 274, 13235–13241.

North, M., Steffen, J., Loguinov, A.V., Zimmerman, G.R., Vulpe, C.D., and Eide,

D.J. (2012). Genome-wide functional profiling identifies genes and processes

important for zinc-limited growth of Saccharomyces cerevisiae. PLoS Genet.

8, e1002699.

Philippi, A., Steinbauer, R., Reiter, A., Fath, S., Leger-Silvestre, I., Milkereit, P.,

Griesenbeck, J., and Tschochner, H. (2010). TOR-dependent reduction in the

expression level of Rrn3p lowers the activity of the yeast RNA Pol I machinery,

but does not account for the strong inhibition of rRNA production. Nucleic

Acids Res. 38, 5315–5326.

Powers, T., and Walter, P. (1999). Regulation of ribosome biogenesis by the

rapamycin-sensitive TOR-signaling pathway in Saccharomyces cerevisiae.

Mol. Biol. Cell 10, 987–1000.

Puig, S., Askeland, E., and Thiele, D.J. (2005). Coordinated remodeling of

cellular metabolism during iron deficiency through targeted mRNA degrada-

tion. Cell 120, 99–110.

Richardson, L.A., Reed, B.J., Charette, J.M., Freed, E.F., Fredrickson, E.K.,

Locke, M.N., Baserga, S.J., and Gardner, R.G. (2012). A conserved deubiqui-

tinating enzyme controls cell growth by regulating RNA polymerase I stability.

Cell Rep 2, 372–385.

Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., and Seraphin, B.

(1999). A generic protein purificationmethod for protein complex characteriza-

tion and proteome exploration. Nat. Biotechnol. 17, 1030–1032.

Shcherbik, N., and Pestov, D.G. (2011). The ubiquitin ligase Rsp5 is required

for ribosome stability in Saccharomyces cerevisiae. RNA 17, 1422–1428.

Simm, C., Lahner, B., Salt, D., LeFurgey, A., Ingram, P., Yandell, B., and Eide,

D.J. (2007). Saccharomyces cerevisiae vacuole in zinc storage and intracel-

lular zinc distribution. Eukaryot. Cell 6, 1166–1177.

Stade, K., Ford, C.S., Guthrie, C., andWeis, K. (1997). Exportin 1 (Crm1p) is an

essential nuclear export factor. Cell 90, 1041–1050.

Stefanovsky, V., Langlois, F., Gagnon-Kugler, T., Rothblum, L.I., and Moss, T.

(2006). Growth factor signaling regulates elongation of RNA polymerase I

transcription in mammals via UBF phosphorylation and r-chromatin remodel-

ing. Mol. Cell 21, 629–639.

Terhune, M.W., and Sandstead, H.H. (1972). Decreased RNA polymerase

activity in mammalian zinc deficiency. Science 177, 68–69.

Toesca, I., Nery, C.R., Fernandez, C.F., Sayani, S., and Chanfreau, G.F. (2011).

Cryptic transcription mediates repression of subtelomeric metal homeostasis

genes. PLoS Genet. 7, e1002163.

Treich, I., Riva, M., and Sentenac, A. (1991). Zinc-binding subunits of yeast

RNA polymerases. J. Biol. Chem. 266, 21971–21976.

Vida, T.A., and Emr, S.D. (1995). A new vital stain for visualizing vacuolar

membrane dynamics and endocytosis in yeast. J. Cell Biol. 128, 779–792.

Warner, J.R. (1999). The economics of ribosome biosynthesis in yeast. Trends

Biochem. Sci. 24, 437–440.

Zhao, H., and Eide, D. (1996). The ZRT2 gene encodes the low affinity zinc

transporter in Saccharomyces cerevisiae. J. Biol. Chem. 271, 23203–23210.

Molecular Cell

RNA Polymerase I Degradation in Zinc Deficiency

Molecular Cell 51, 105–115, July 11, 2013 ª2013 Elsevier Inc. 115