Inducible Gene RNR3 in Saccharomyces cerevisiae

Abstract

Ribonucleotide reductase is an essential enzyme that catalyzes the rate limiting step for production of the deoxyribonucleotides required for DNA synthesis. It is encoded by three genes, RNRI, RNRZ and RNR3, each of which is inducible by agents that damage DNA or block DNA replication. To probe the signaling pathway mediating this DNA damage response, we have designed a general selection system for isolating spontaneous trans-acting mutations that alter RNR3 expression using a chromosomal RNR3-URA3 transcriptional fusion and an RNR3-lacZ reporter plasmid. Using this system, we have isolated 202 independent trans-acting crt (constitutive RNR3 transcription) mutants that express high levels of RNR3 in the absence of DNA damaging agents. Of these, 200 are recessive and fall into 9 complementation groups. In some crt groups, the expression of RNRI and RNRZ are also elevated, suggesting that all three RNR genes share a common regulatory pathway. Mutations in most CRT genes confer additional phenotypes, among these are dumpiness, hydroxyurea sensitivity, temperature sensitivity and slow growth. Five of the CRT genes have been identified as previously cloned genes; CRT4 is TUPI, CRTS is POLI/CDC17, CRT6 is RNR2, CRT7 is RNRI, and CRT8 is SSN6. crt6-68 and crt7-240 are the first ts alleles of RNRZ and RNRI, respectively, and arrest with a large budded, cdc terminal phenotype at the nonpermissive temperature. The isolation of crt5-262, an additional cdc allele of POLI/CDCI7, suggests for the first time that directly blocking DNA replication can provide a signal to induce the DNA damage response. crt2 mutants show a defect in basal level expression of RNRI-lacZ reporter constructs. These are the first mutants isolated in yeast that alter the regulation of DNA damage inducible genes and the identification of their functions sheds light on the DNA damage sensory network. T HE DNA damage sensory response pathway is one of several global regulatory networks that coordinately control the expression of a large number of genes in response to environmental stress. These capacities are central to an organism’s ability to successfully adapt to life threatening environmental conditions and to duplicate its genetic material with the highest fidelity. Both procaryotes and eukaryotes are able to sense and respond to DNA damage. However, this sensory network is well understood only in the procaryote Escherichia coli. Treatment of E. coli with agents that damage DNA or block replication causes the appearance of a set of physiological responses that include the induction of DNA repair processes, mutagenesis, and induction of lysogenic bacteriophage (WALKER 1985). These processes have collectively been called the SOS response because at least some of them appear to promote cell survival. The molecular mechanism of this coordinately regulated response involves the proteolytic inactivation of a common repressor, the LexA protein, by an activated form of the RecA protein. The RecA protein can become activated to facilitate proteolysis by binding to singleGenetics 131: 851-866 (August, 1992) stranded DNA, a possible damage signal. The functions of several SOS regulated genes are known and include excision repair (uvrAB) (KENYON and WALKER 198 l), recombinational repair (recA) (MCENTEE 1977), SOS repression (lexA) (LITTLE and HARPER 1979), mutagenesis (umuCD, mucAB) (ELLEDGE and WALKER 1983a,b), inhibition of cell septation, i.e., cell cycle arrest ($A) (HUISMAN, D’ARI and GEORGE 1980), and possibly site-specific recombination (himA) (MILLER, KIRK and ECHOLS 198 1). Eucaryotic organisms display similar cellular responses to DNA damage including cell cycle arrest (HITTLEMAN and RAO 1974; KUPIEC and SIMCHEN 1985), increased capacity to carryout deoxyribonucleotide synthesis (LAMMERS and FOLLMANN 1986; ELLEDGE and DAVIS 1990), transcriptional induction of genes involved in DNA replication and DNA damage repair pathways (see below), and increased mutagenesis frequency (SARASIN and BENOIT 1980; MITCHEL and MORRISON 1986). In the yeast Saccharomyces cerevisiae, a number of damage inducible genes have been identified. Known functions of DNA damage inducible genes include excision repair [RAD2 (ROBINSON et 852 Z. Zhou and S . J. Elledge al. 1986)], recombinational repair [RAD54 (COLE et al. 1987)], alkylation repair [MAGI (CHEN, DERFLER and SAMSON 1990)], photoreactivation [PHRl (SEBASTIAN, KRAUS and SANCAR 1990)], the ubiquitin proteolysis pathway [UBZ4 (TREGER, HEICHMAN and MCENTEE 1988)], and a number of genes encoding S phase-specific, cell cycle-regulated activities involved in nucleotide synthesis [RNRl , RNR2, RNR3 and CDC8 (ELLEDCE and DAVIS 1987; HURD, ROBERTS and ROBERTS 1987; ELLEDCE and DAVIS 1990)] and DNA synthesis [POLl/CDCl7 (JOHNSTON et al. 1987), and CDC9 (BARKER, WHITE and JOHNSTON 1985; PETERSON et al. 1985)l. The functions of two sets of genes identified on the basis of increased transcription in response to DNA damage, the DIN (RUBY and SZOSTAK 1985) and DDR genes (MCCLANAHAN and MCENTEE 1984), remain largely unknown. The inducibility of cell cycle regulated genes is thought to be due to the temporal compartmentalization of the cell cycle in which much of the capacity to synthesize DNA is restricted to S phase. The induction of these genes in response to the stress of DNA damage is thought to produce a metabolic state resembling S phase that facilitates DNA replicational repair processes (ELLEDGE, ZHOU and ALLEN 1992). Among DNA damage inducible genes, regulation of the RNR gene family has been studied most extensively. Ribonucleotide reductase catalyzes the reduction of ribonucleotides to deoxyribonucleotides, the first and rate limiting step in the pathway for the production of the deoxyribonucleotides needed for DNA synthesis. It is an enzyme of structure a&. In yeast, the small subunit is encoded by RNRP (ELLEDGE and DAVIS 1987; HURD, ROBERTS and ROBERTS 1987), and the large subunit is encoded by two homologous genes, RNRl and RNR3 (ELLEDCE and DAVIS 1990). Hydroxyurea (HU) functions as a potent inhibitor of RNR activity by quenching the free radical in the active site of the small subunit (HARDER and FOLLMANN 1990). The large subunit contains two sites for the allosteric regulation of the enzyme activity: one site controls the substrate specificity, while the other site controls the overall activity of the enzyme (REICHARD 1988). Both RNRl and RNRB are essential for mitotic viability, while disruption of RNR3 has no obvious phenotype under normal growth conditions. However, RNR3 encodes an active protein since increased gene dosage of RNR3 is capable of complementing a null allele of RNRl (ELLEDGE and DAVIS 1990). R N R l , RNRB and RNR3 are all inducible at the level of transcript accumulation by agents that block DNA synthesis through nucleotide depletion such as HU and methotrexate, or by agents that damage DNA, such as UV light, methyl methanesulfonate (MMS) and 4-nitroquinoline-1 -oxide (4-NQO) (ELLEDCE and DAVIS 1989a; HURD and ROBERTS 1989; ELLEDGE and DAVIS 1990). RNRl is inducible 5-fold, RNR2 is inducible 25-fold, and RNR3 can be induced 100-500-fold. Under normal growth conditions the RNR3 transcript is undetectable and is therefore under very tight control. Why an organism would retain the function and tight regulation of RNR3 remains a mystery. RNR3 may allow the induction of a ribonucleotide reductase with altered regulatory properties that confers a selective advantage on the cell. A detailed deletion analysis of the RNR2 regulatory region has implicated both a positive and a negative regulatory element in this response and identified a 70-bp fragment termed the DRE (DNA damage responsive element) that can confer some DNA-damage inducibility upon a heterologous promoter. A number of proteins were shown to bind to this element including the Rap1 protein. However, it has not been determined whether any of these are essential for the response (ELLEDGE and DAVIS 198913). Although RNRl and RNR3 are inducible, none of these proteins are bound by DNA from the promoters of RNRl or RNR3 (Z. ZHOU and S. J. ELLEDGE, unpublished) suggesting that either they do not share transcription factors, or that any common factors have not yet been detected. In all, this detailed analysis has revealed the complex nature of the RNRB regulatory region and suggests that promoter dissection alone may not be the most general approach to understanding the DNA damage induction pathway in its entirety. Very little is known about the mechanisms employed to sense and respond to DNA damage in eukaryotes. Although several different types of DNA damage and replication blocks have been found to induce multiple genes, several outstanding questions remain including: (1) Are many different damage signals sensed by the sensor separately or are they transformed to a single form prior to detection? (2) Is nucleotide depletion itself a signal, or does induction result from the subsequent block in DNA synthesis? (3) Are all DNA damage inducible genes regulated by a common mechanism or are there different regulatory pathways such as for recA and ada in E. coli (WALKER 1985)? Mutations in the genes encoding signal sensing and transducing proteins would facilitate the identification of the components and mechanisms of these pathways. T o begin isolation of these mutants, we have chosen RNR3 as a target gene due to the fact that it is nonessential and highly inducible by DNA damaging agents. Here we report the isolation and the characterization of spontaneous mutations that cause constitutive expression of RNR3 (crt mutants) in the absence of DNA damage, and discuss the possible roles of the CRT genes in the damage signal sensing and transducing pathways. Mutants Constitutive for RNR3 853 MATERIALS AND METHODS Media and chemicals: Yeast minimal medium contains 0.67% yeast nitrogen base without amino acids (Difco Laboratories, Detroit Michigan), 2% glucose and 2% BactoAgar (difco) is added for solid media. Synthetic complete media (SC) is minimal medium supplemented with various amino acids and bases as described (SHERMAN, FINK and LAWRENCE 1979), as was YPD media. 5-Fluoroorotic acid (5-FOA) was purchased from PCR Incorporated, Gainsville, Florida. 5-FOA plates are prepared as described by ROSE, WINSTON and HEITER (1990). MMS was purchased from Eastman Kodak Co. (Rochester, New York). HU and 5bromo-4-chloro-3-indolyl-/3-~galactopyranoside (X-GAL) were purchased from Sigma Chemical Co. (St. Louis, Missouri). Restriction enzymes, T 4 polymerase, Klenow fragment of DNA polymerase I, and DNA ligase were purchased from New England Biolabs and used under the conditions suggested by the supplier. Strains and plasmids: Yeast strains and plasmids used in this study are listed in Table 1. All crt mutants were isolated from strains Y206 and Y207. Strain H17ClA1 was provided by LELAND HARTWELL. RTY418 and pFW 1-1 (TUPl) were gifts of ROBERT RUMBLY (WILLIAMS and TRUMBLY 1990). pS63 15(SSN6) was kindly provided by JANET SCHULTZ and MARIAN CARLSON. E. coli strain JM107 (YANISCH-PERRON, VIEIRA and MESSING 1985) was used as the host strain for plasmid constructions and amplification. The construction of other plasmids and strains are described below. Plasmid constructions: The 880-bp Sau3Al-PvuII fragment of the RNR3 promoter was cloned into BamHI-EcoRV digested pBS KS+ (Stratagene, San Diego, California) to create pZZ1. The PvuII site in the RNR3 fragment was located at codon 61 of the RNR3 coding sequence (YAGLE and MCENTEE 1990) and the Sau3A1 site regenerated a BarnHI site upon ligation to pBS KS+. Introduction of this fragment into this polylinker was necessary to adjust the reading frame of RNR3 to match the lac2 fragment at the XhoI site. The BamHI-XhoI RNR3 fragment was excised from pZZl and cloned into BamHI-XhoI cleaved pNN407 (ELLEDCE and DAVIS 1989b), a URA3 CEN4 ARSl RNR2lac2 fusion vector, replacing the RNR2 sequences to produce pZZ2. pZZ2 contains an RNR3-lac2 protein fusion. pZZ13, a HIS3 version of pZZ2, was created by excising the RNR3lac2 gene from pZZ2 on an EcoRI-NsiI(b1unt) fragment and inserting this fragment into EcoRI-NarI(blunt) cleaved PUN90 (ELLEDCE and DAVIS 1988). The designation (blunt) after a restriction site name indicates that in this reaction, that site was filled-in or excised to make flush ends by use of klenow fragment of DNA polymerase I and dNTPs. pNN405 (ELLEDCE and DAVIS 1989a) is a CYCl-lac2 fusion driven by regulatory elements from the RNR2 upstream region on a 2-rm URA3 plasmid. URA3 on pNN405 was replaced with HZS3 by inserting a 2.5-kb SmaI fragment from pJA9 (J. B. ALLEN and S. J. ELLEDCE, unpublished) carrying HIS3 and Tn5 neo for selection into StuI cleaved pNN405 to create pZZ18. pZZ 19 has sequences from the upstream region of RNRl conferring DNA damage inducibility on a CYCl-lac2 fusion gene and was constructed as follows. A 242 bp EcoR1 (blunt)DdeI(blunt) fragment of the RNRl upstream region (S. J. ELLEDCE, unpublished) was ligated into SmaI cleaved pBS KS+ to make pSE816. These RNRl sequences were removed from pSE8 16 by digestion with BamHI and XhoI and ligated into BamHI-XhoI cleaved pLG312 ASS (ELLEKE and DAVIS 1989b) to create pSE836. The RNRl fragment in pSE836 has the same orientation relative to the CYCl TATA as it does in the native RNRl promoter. URA3 on pSE836 was replaced with HIS3 by inserting a 2.5-kb SmaI fragment from pJA9 carrying HIS3 and Tn5 neo for selection into StuI cleaved pSE836 to create pZZ 19. The construction of pZZl1, which contains the transplacement cassette for replacing the RNR3 gene with the RNR3-URA3 transcriptional fusion, was quite complicated and rather than describing the circuitous route of construction, we will describe the segments in linear order. At nucleotide 1 is a 52-bp oligonucleotide of sequence TAGCAGCAACAATAAATCTCGAG. KpnI and XhoI sites are underlined, respectively. This sequence directly precedes the start codon of RNR3 translation (YAGLE and MCENTEE 1990). Adjacent to the XhoI site is a 900-bp PstIHindIII URA3 fragment. The PstI site in URA3 is 17 bp before its start codon. Both the PstI and XhoI sites were made blunt and destroyed in the ligation. Adjacent to the HindIII site is a 12-bp sequence from pBS KS+, GGTGGCGGCCGC, which contains a NotI site. Adjacent to the NotI site is a 1.5-kb BglII fragment containing TRPl. This is the genomic TRPl BglII fragment with the EcoRI site in the TRPl promoter filled-in with klenow and dNTPs (P. HEITER, unpublished). Adjacent to this fragment is a 30 bp hybrid polylinker fragment containing the restriction enzyme cleavage sites Sad, EcoRI, XbaI, MluI and SalI. Adjacent to the SalI site is a 1 .O-kb SalI-KpnI fragment from the RNR3 gene. The remaining fragment is the entire 3.0 kb pBS KS+ plasmid cleaved with KpnI. The 5‘ RNR3URA3-TRPl-RNR3 3’ sequences can be liberated from this plasmid by cleavage with KpnI and used for transplacement into yeast. pZZ17 was created by inserting the EcoRI-Aut11 (blunt) LEU2 fragment from YIplacl28 (GIETZ and SUGINO 1988) into EcoRI-Not1 (blunt) cleaved pZZl1, replacing the TRPl gene. The LEU2 gene on YIplacl28 is engineered to remove the EcoR1 and KpnI sites. pZZ16, the plasmid used to create the 5’ ura3 deletion, was made in several steps. First the 1.1 -kb URA3 gene on a HindIII-SmaI fragment was cloned into HindIII-SmaI cleaved pUC 19 (YANISCH-PERRON, VIEIRA and MESSING 1985) to create pSE272. A 2-kb NcoI (blunt)-EcoRI fragment from pZZl1 containing 500 bp of the 5’ end of URA3 and the entire TRPl gene was ligated into PstI (blunt)-EcoRI cleaved pSE272 to create pZZ15. This created a 200-bp deletion in the URA3 gene between PstI and NcoI. The StuI site in the 3’ region of TRPl was destroyed by cleaving pZZ15 with PstI, making flush with T4 polymerase and the dNTPs, and religating to produce pZZ16. This PstI site overlaps the StuI site by 2 bp and both sites are destroyed in this series of reactions. The remaining StuI site in the 3’ end of the URA3 sequence is unique and can be used to target the plasmid to the URA3 locus. Construction of yeast strains: Y200, containing the RNR3-URA3 transcriptional fusion replacing the RNR3 gene, was created by transforming XS955-36B (a gift from D. SCHILD) with KpnI cleaved pZZl1 and selecting for Trp’. Transformants were chosen that were Urabut became Ura+ when grown on SC-Ura plates with 50 mM HU. Southern analysis was performed to confirm that the transplacement event had occurred and that the RNR3 gene had been replaced with the RNR3-URA3 fusion linked to TRPl. Y202 was constructed in two steps. First, pZZ16, containing ura3-Al00 (APstI-NcoI) linked to TRPl, was cleaved with StuI which cuts in the 3’ URA3 sequences and transformed into Y201 selecting for Trp+ colonies. These were grown in YPD liquid for overnight and streaked on 5-FOA plate to select for Uracolonies in which the intact URA3 and the TRPl genes were looped out and a deleted ura3GGTACCCTTGAAATAAATAT-GACAAGCAAGAAa 5 4 Z. Zhou and S. J. Elledge

8 Figures and Tables

Cite this paper

@inproceedings{Zhou2002InducibleGR, title={Inducible Gene RNR3 in Saccharomyces cerevisiae}, author={Zhengyang Zhou}, year={2002} }