Defining Error-Free Postreplication Repair In Saccharomyces cerevisiae
A thesis submitted to the College of Graduate Studies and Research in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Department of Microbiology and Immuriology University of Saskatchewan
Stacey D. Broomfield, B.Sc.
Winter 2000
O Copyright Stacey D. Broornfield, 2000. Al1 rights reserved. National Library Bibliothèque nationale 1+1 of Canada du Canada Acquisitions and Acquisitions et Bibliographic Se~ices services bibliographiques 395 Wellington çtreet 395, nie Wellington Ottawa ON KIA ON4 Ottawa ON K1A ON4 Canada Canada
The author has granted a non- L'auteur a accordé une licence non exclusive licence aIlowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or sell reproduire, prêter, distribuer ou copies of this thesis in microform, vendre des copies de cette thèse sous paper or elertronic formats. la forme de microfiche/fFlm, de reproduction sur papier ou sur foxmat électronique.
The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or othenvise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. Permission To Use
In presenting this thesis in partial fulfillment of the requirements for a
Postgraduate degree from the University of Saskatchewan, 1agree that the
Libraries of this University may make it freely available for inspection. 1further agree that permission for copying of this thesis in any manner, in whole or in part, for scholarly purposes may be granted by the professor or professors who supervised my thesis work or, in their absence, by the Head of the Department or the Dean of the College in which my thesis work was done. It is understood that any copying or publication or use of this thesis or parts thereof for financial gain shall not be allowed without my written permission. It is also understood that due recognition shall be given to me and to the University of Saskatchewan in any scholarly use which may be made of any material in my thesis.
Requests for permission to copy or make other use of material in this thesis in whole or in part should be addressed to:
Head of the Department of Microbiology and Immunology
Health Science Building, 107 Wiggins Road,
University of Saskatchewan
Saskatoon, Saskatchewan
Canada S7N 5E5 Abstract
The mms2-1 yeast mutant was isolated by its sensitivity to the DNA
alkylating agent methyl methanesulfonate (MMS) as part of a screen to identify
genes respowible for repair of DNA alkylation damage. 1 cloned the MMS2
gene by complementation of the mms2-1 mutant for MMS sensitivity and
determined its nucleotide sequence. The deduced Mms2 protein shares strong sequence similarity with ubiquitin-conjugating enzymes (Ubcs), but Mms2 lacks
in vitro Ubc activity. Unlike the original mms2-l mutant, which is only sensitive
to MMS, the mms2 disruption mutant displays sensitivity to both MMS and UV radiation. Another Ubc mutant that displays both UV and MMS sensitivity is
ubc2/rad6 which is essential for both postreplication repair (PRR) and muhgenesis. DNA damage that blocks normal replication is lethal. PRR is a mechanism that allows replication to proceed in the presence of damage and for
repair to be effected thereafter.
The rad6 and rad18 mutations are epistatic to mms2, suggesting that
MMSZ functions in the RAD6 pathway. Interestingly, the mms2 mutation
increases both spontaneous and UV-induced mutagenesis, is synergistic with
rev3 for UV and MMS sensitivity, and its mutagenesis effect is completely
dependent on the REV3 gene. Based on these observations, it is proposed that
MMSZ plays a role in error-free PRR parallel to REV3-mediated mutagenesis
within the RAD6 pathway.
Rad6 has many cellular functions that are not related to PRil, i~ciudkg
telomeric silencing, sporulation and protein degradation. To determine whether or not Mms2 functions as an accessory protein for Rad6 or if it is specific for
PRR, 1 tested the effect of the mms2 mutation on each of the above cellular processes. mms2 displayed a moderate defect in sporulation, with a minor effect on protein degradation and no effect on teelomeric silencing. Hence, 1 have determined that Mms2 is involved in error-free PRR, but its role in other Rad6 mediated processes is yet to be determined.
Several proteins or domains, including PCNA (Po130), Rad30, Rad5 and the N-terminus of Rad6, are irnplicated in the error-free PRR pathway. We perfomed epistasis analysis to ask whether or not these genes and mms2 act in the same error-free subpathway. None of the above mutations exhibited epis ta tic rela tionships, indica ting tha t the gene tic interactions among these genes are complica ted.
The SRS2 gene encodes a helicase whose exact function is unknown.
Alleles of srs2 were identified based on their ability to suppress the extreme UV and MMS sensitivity of rad6 and rad18 mutants. 1 wanted to assess the effect of the srs2 nul1 mutation on the various mutants defective in PRR. 1 found that sr52 is epistatic to al1 PRR mutations, regardless of error-free and error-prone involvement, since the srs2 mutation was able to suppress the MMS sensitivities of po130-46, rad5, pol30-46 rad5, rad30, mms2 and rev3. These genetic relationships indicate that the PRR pathway is dependent upon the unknown activity of the
Srs2 helicase.
Two human homologues of MMS2 were isolated. CROC-IS was isolated due to its ability to transactivate the c-fos promoter. hMMS2 was isolated in a homology screen using the deduced Mms2 sequence. The 145 amino acid
iii cDNA clone shares 50.4% sequence identity to Mrns2. Since sequence homology alone cannot determine if the two enzymes evolved with similar functions, 1 cioned hMMS2 under the conhol of a yeast inducible promoter and introduced it into a mms2 mutant to assay for functional complementation. hMMS was able to complernent the MM5 and UV sensitivities, as well as mutator phenotypes of the mms2 mutant. CROC-1B is unable to complement any of the mms2 defects, unless its N-terminal domain is removed.
My research with yeast and hurnan MMS2 has helped to clariQ the PRR pathway and to define a novel Ubc-like protein family. This study also aided the recent advances in the biochemistry of a novel Ubcl3-Mms2 complex formation, its unique ubiquitin chah assembly and its interactions with RING finger proteins. In addition, discoveries made in this study have stimulated Our laboratory as well as other laboratories to focus on the roles of the Ubcl3-Mms2 complex in human diseases and cancer. IiMMS2 and CROC4 have been implicated in the progression of cells from a pre-immortal to immortal state, in cellular differentiation, in tumorigenesis and in error-free PRR. Further investigations of this complex may prove to be useful in cancer prevention and therap y. Acknowledgments
1would like to begin by thanking my supervisor, Dr. Wei Xiao. Dr. Xiao has provided a supportive learning environment and invaluable guidance during a very busy time in his career. His devotion, enthusiasm and encouragement has been a motivating force throughout my studies, and is greatly appreciated. 1would also like to thank the members of my supewisory cornmittee, Dr. Susan Laferte, Dr. Hamy Deneer, and Dr. Sean Hernmingsen, for offering support and advice. A special thanks to Dr. Dan Gietz, my extemal examiner from the University of Manitoba. Dr. Gietz was a wonderful addition to my defense proceedings.
A very important thank-you is extended to the people in the laboratory of
Dr. Xiao. Our laboratory has always contained talented individuals that are a pleasure to work with. Thanks to Treena Swanston, Barbara Chow, Todd
Hryciw, Mahmood Chamankhah, Yule Liu, Janelle Franko, Sonya Bawa,
Carolyn Ashley, Leslie Sarbour, Landon Pastushok, Yu Zhu, Parker Anderson and Michelle Hama, for making the workplace fun. 1must Say a special thank- you to Todd Hryciw, colleague and friend. Todd is a friend with whom 1 shared excitement, frustration, laughs, ideas, and al1 the trivial problems life offers.
1greatly appreciate the financial support provided by College of
Medicine Graduate scholarships and the University of Saskatchewan Graduate scholarships.
Finally, 1must thank my family. Thank you for your self-less cornmitment, patience and unconditional love. My gratitude camot be expressed in words.
Table of Contents:
PERMISSION TO USE
ABSTRACT
ACKNOWLEDGMENTS
DEDICATION
TABLE OF CONTENTS vii LIST OF TABLES
LIST OF FIGURES xii
LIST OF ABBREVIATIONS xiv CHAPTER ONE: INTRODUCTION
1.1. Protecting Our Genome.
1.2. DNA Damage. 1.2.1. DNA Damage. 1.2.2. DNA Damage: Biological Consequences. 1.2.3. Endogenous DNA Damage. 1.2.4. Environmental DNA Damaging Agents. 1.2.4.l. Physical Agents. 1.2.4.2. Chernical Agents.
1.3. How the CeIl Copes with DNA Damage. 1.3.1. Ce11 Cycle Controls. 1.3.2. DNA Damage Repair.
1.4. DNA Repair Pathways. 1.4.1. Direct Reversa1 Of DNA Damage. 1.4.2. Excision Of DNA Damage. 1.4.2.1. Base Excision Repair. 1.4.2.2. Nucleotide Excision Repair. 1.4.3. Recombina tion Repair.
1.5. DNA Darnage Tolerance. 1.5.1. SOS Response Ln E. coli 1.5.2. Postreplication Repair in Eukaryotes
1.6. Postreplication Repair In Saccharomyces cerevisiae 1.6.1. Recombination Repair In DNA Damage Tolerance 1.6.2. Assessing The Genes Comprising PRR.
vii 1.6.2.1. RAD6-RAD18 Cornplex: DNA Damage Recognition? 36 1.6.2.2. Rad6: A Ubiquitin-conjugating Enzyme. 37 1.6.2.3. Translesional Replication. 40 1.6.2.3.1. The Mutagenic Response. 41 1.6.2.3.2. Error-Free Translesional Bypass: RAD30 . 42 1.6.2.4. Clarifymg Error-Free PRR. 45 1.6.2.4.1. Suppressors Of rad6 and rad1 8 Mutants. 45 1.6.2.4.2. Other Candidate Genes In Error-Free PRR. 47
1.7. MMS2: Rationale For This Projed. 51
CHAPTER IWO: MATERIALS AND METHODS. 54
2.1. Yeast Genetics. 2.1.1. Yeast Strains And Ce11 Culture. 2.1.2. Yeast Transformation. 2.1.3. Yeast Plasmid Extraction. 2.1.4. Yeast Genomic DNA Isolation. 2.1.5. Screening A Yeast Genomic Library. 2.1.6. Co-Segregation Test. 2.1.7. Drugs And Special Media. 2.1.8. Ce11 Killing. 2.1.9. Spontaneous Mutagenesis Assay . 2.1.10. UV-Induced Mutagenesis Assay. 2.1.1 1. Assessing Telomeric Silencing. 2.1.12. Protein Degradation Assay. 2.1 .l3. Sponilation Assaÿ.
2.2. Molecular Biology Techniques. 2.2.1. Bacterial Culture And Storage. 2.2.2. Plasmid DNA Isolation (Mini-Prep). 2.2.3. Large Scale DNA Isolation (Maxi-Prep). 2.2.4. Making Competent Cells For Transformation. 2.2.5. Bacterial Transformation. 2.2.5.1. Chernical Method. 2.2.5.2. Eledroporation Method. 2.2.6. DNA Sequencing. 2.2.7. PCR Amplification. 2.2.8. Agarose Gel Electrophoresis. 2.2.9. Isolation Of DNA Fragments From Agarose Gels. 2.2.10. Ligation. 2.2.11. Probe Labelling. 2.2.12. Southern Hybridization. 2-2-13.Plasmids And Plasrnid Construction. CHAPTER THREE: RESULTS.
3.1. Molecular Cloning Of MMSZ. 3.1 .l. Isolation of Clone From Genomic Library. 3.1.2. Mapping The YCpMî Clone.
3.2. Initial Analysis Of -2. 3.2.1. MM52 1s Allelic To mms2-1 3.2.2. Phenotypes Of mms2A. 3.2.3. Sequence Analysis Of MMSZ. 3.2.3.1. Analysis Of Nudeotide Sequence. 3.2.3.2. Analysis Of Amino Atid Sequence.
3.3. The mms2-l Allele. 3.3.1. Cloning And Sequencing Of mms2-1. 3.3.2. mms2-1: A Partial Loss of Function Mutation.
3.4. MMS2: Genetic Interactions. 3.4.1. MMS2 Belongs To The RAD6 Epistasis Croup. 3.4.2. MMS2 1s Not A Member Of NER or Recombination Repair. 3.4.3. MMS2 And RE V3 Mutagenesis. 3.4.4. rad6,,-, is Epistatic to mms2A. 3.4.5. Epistatic Analysis Within Error-Free PRR
3.5. Mms2: Acressory Protein To Rad6 Or A Enzyme Exclusive To PRR? 3.5.1. Spordation. 3.5.2. Telomenc Silencing. 3.5.3. Protein Degradation Via The N-End Rule. 3.5.4. Associated Functions Of Mms2
3.6. SRS2: Involvement In Error-Free PRR. 3.6.1. Epistasis Between srs2 and mms2. 3.6.2. UV-Induced Mutagenesis. 3.6.3. Epistasis Between srs2 And The Other Members Of Error-Free PRR. 3.6.4. SRS2 And Error-Free Translesional Bypass.
3.7. MMS2 Human Homologues. 3.7.1. Identification Of hMMS2. 3.7.2. Fundional Complementation Of The mms2 Mutant By h2MMS2 hdCROC-1
CHAPTER FOUR: DISCUSSION.
4.1. Redefining The Rad6 Pathway. 4.1.1. MMS2 1s Part Of The RAD6 PRR Pathway. 4.1.2. MMS2 And The Error-Free PRR Pathway. 4.1 -3. Organization Of Error-Free PRR.
4.2. Mms2 In Sporulation and Protein Degradation
4.3. MMS2 And UBC13-
4.5. The Role Of Srs2 In DNA Damage Tolerance.
4.6. Evolutionary Conservation.
REFERENCES List of Tables
Table
Sncchromyces cerevisiae strains
Oligonudeotide sequences
Plasmid constructs
Spontaneous mutation rates of S. cerevisiae strains
UV-induced mutagenesis of S. cereuisiae strains
Sporulation assay
The srs2 mutation and UV-induced mutagenesis
8 hMMS2: Complementation of the mms2 mutator phenotype 140 List of Figures
Fimire Page
1 Proposed models for error-free translesional bypass 34
2 The ubiquitin pathway 38
3 Postreplication repair overview 51
4 Mapping and disruption of the MM52 gene
5 Nucleotide sequence of MMS2
6 Sequence alignment of Mms2 and other Ubcs in S. cerevisiae 90
7 Cornparison of the MMç sensitivities of the mms2 null mutant and the nzms2-1 allele
8 Over-expression of the rnrns2-2 allele complements the MMS sensitive phenotype of the mms2 null mutant
9 Assessing the genetic interactions of mms2 in response to UV-irradiation
10 Assessing the genetic interactions of mms2 in response to MMS 98
11 mm2 is synergistic to rev3 with respect to both W and MMS sewitivity
12 rnd6,,, is epistatic to mms2A with respect to both UV 105 and MMS sensitivity
13 Epistasis analysis between rad5 and mms2 within error-free PRR 108
14 Epistasis analysis between mms2 and poZ30-46 109
15 Assessing the involvement of MMS2 in translesional bypass 110
16 The mms2 mutation does not alter telomere silencing 115
17 The relative P-galactosidase activity, as a measure of ubiquitination 117 via the N-end rule
18 srs2 is epistatic to mms2 121
19 The effed of the srs2 mutation on the spontaneous mutation 122 rate of mms2
20 SRS2 and the mutagenic response 124
21 Suppression of the mms2 rm3 MA6sensitivity by srs2 125
22 The effect of the srs2 mutation on the members of error-free PRR 131
23 Epistasis analysis between rad30 and srs2 132
24 Nucleotide and deduced amino acid sequences of the hMMS2 cDNA 134
25 Sequence alfigrunent of Mms2 and its human homologues 136
26 Complemention of mms2 by hMMS2 139
27 Truncated CROC-1B is able to complement the mm2 mutant 141
28 Complementation and homology oveMew of mms2 with hMMS2 and CROC4 List of Abbreviations aa amino acid A~P ampicilin AP site apurinic or apyrimidinic site ATP adenosine 5'-triphosphate P-pal P-galactosidase BER base excision repair bp base pair cDNA copy DNA CE' calf intestinal phosphatase CS Cockayne's syndrome C-terminus carboxy terminus A deletion mutation (nul1 mutation) DMÇO dimethyl sulfoxide DNA deoxyribonudeic aad dNTr 2'-deoxynucleotide 5'triphosphate DSBs double stranded breaks dsDNA double stranded DNA El ubiquitin adivating erqpe E2 ubiquitin-conjugating enzyme (Ubc) E3 ubiquitin protein ligase E. coli Escherichia coli ESTs expressed cDNA sequence tags EDTA ethylenediaminetetraacetic acid EtBr ethidiurn bromide FQY 2'6-diamino-4-hydroxy -5-formamidopyrimidine FOA 5-fluoro-orotic acid G1 phase gap 1 phase of the ce11 cycle G2 phase gap 2 phase of the ce11 cycle GG-NER global genome NER EIGS Human Genome Sciences Inc. database mcc hereditary nonpolyposis colorectal cancer IR ionizing radiation kb kilobase pair or kilobase kDa kilodalton M phase rnitosis phase of the ce11 cycle MAT mathg type locus MCS multiple doning site 3MeA 3-methyladenine 2MeC 2-methylcytosine 7MeG 7-methy lguanine MMS methyl methanesulfonate MNNG N-methyl-W-nitro-N-nitrosoguanidine mRNA messenger RNA MW molecular weight
xiv N-end rule amino end de N-tenninus arnino terminus NER nucleotide exasion repair nt nudeotide OD optical density 06MeG 06-methylguanine 04MeT 04-methylthymine 8-0~0-G 7f8-Dihydro-8-oxo-guanine ORF open reading frame PCNA proliferating ce11 nuclear antigen PCR polyrnerase chah reaction PEG polyethylene glycol PRR postreplication repair Red" activated Red RNA ribonucleic acid S phase synthesis phase (genome duplication) of the ce11 cycle SCE sister chromatid exchange S. cerevisiae Saccharomyces ceratisiae SGD Saccharomyces Genome Database ÇN1 nudeophilic substitution mechanism 1 SN2 nudeop hilic substitution m-zchanism 2 SSB single stranded breaks ssDNA single stranded DNA TCR-NER transcription-coupled repair of NER ts temperature sensitive Ub ubiquitin Ubc ubiquitin conjugating enzyme UmuDf carboxyl terminal domain of UmuD UV ultraviolet radiation w wild type XP xeroderma pigmentosum YCP yeast centromeric plasmid ~EP yeast episome plasmid CHAPTER ONE: INTRODUCTION
1.1. Protecting Our Genome
DNA is not inert. The consequences of modifying the coding properties of DNA can Vary from beneficial to detrimental. If genomic DNA were completely resistant to alterations, organisms would not be able to evolve. Conversely, the detrimental effect of massive alterations to genomic DNA, or mutations in genes that are involved in essential cellular functions, is ce11 death or various diseîses including cancer.
The link between mutagenesis and cancer has sparked the interest of the public and of the medical community. Research initiatives have been aimed at identifying different agents that destabilize the genome, with the purpose of reducing exposure to these DNA-damaging agents. In the process of determining what agents are harmful, researchers have gained understanding in how cells manage stress to the genome. Part of the stress management requires the activity of various DNA damage repair pathways, whose purpose is to reverse the effects of damage incurred by genomic DNA from endogenous and environmental sources.
The importance of DNA damage repair is apparent from both an evolutionary and a medical perspective. The evolutionary conservation of DNA damage repair pathways indicates the cellular importance of repairing damaged
DNA. From a medical standpoint, the clinical relevance of DNA repair processes is observed in individuals that lack such pathways. Patients with diseases such as xeroderma pigrnentosum, Cockayne syndrome, and ataxia- telangiectasia have defects in particular DNA repair pathways. These patients display syrnptoms that include poor growth, neurological abnormalities, increased susceptibility to cancer, photosensitivity, and immunodeficiency
(Friedberg et al., 1995). Familial forms of cancer have also been linked to defects in DNA repair pathways. For example, the majority of hereditary nonpolyposis colorectal cancer (HNPCC) cases are associated with loss of the DNA mismatch repair pathway (Modrich, 1994). These correlations between dysfunctional
DNA repair pathways and disease elicited hope that defects in DNA repair could serve as a marker in the prediction of diseases like cancer. Unforhuiately, in the field of cancer research, this has not been fulfilled. Genetically, cancer cells are characterized by gross chromosomal alterations, rarely with defects in
DNA repair enzymes (Lengauer et nl., 1998). What do we know about carcinogenesis? The ce11 contains DNA repair pathways, equipped to remove genetic damage caused by metabolic byproducts or by replication errors. We are aware of environmental mutagens that increase the incidence of mutagenic events, thus increasing the chance of altering the biochemistry within individual cells. Ln multicellular organisms, the correlation between mutagens and carcinogenesis is found in the molecular characterization of cancerous cells. Mutations in the p53 gene are the most notorious alteration found in cancer cells, but there are many more examples of gene mutations linked to carcinogenesis (reviewed in Hemrninki, 1993). The process of mutagenesis is dependent upon whether the affected cell's ability to resolve the damage is compromised. Knowing this, cancer prevention may rely on determining how the adducts translate into cancer. 1.2. DNA Damage
1.2.1. DNA Damage
The fondation of life is contained within the DNA-protein complexes that occupy each living cell. The instructive component of the DNA molecule is contained in the pattern of nitrogen bases (adenine, thymine, cytosine, and guanine); each base is held within the sugar-phosphate backbone of the polynucleotide chah tluough an N-glycosylic bond (Voet & Voet, 1990). DNA helix formation relies on the proper hydrogen bonding between complementary bases on opposite polynucleotide strands. Complementarity of bases matches cytosine witi? guanine via three hydrogen bonds, and adenine with thymine via two hydrogen bonds.
The genetic integrity of each species relies upon the maintenance of the
DNA structure, which is at risk since DNA is subject to reactions that can change its chernical formula. Chernical modification can occur at many parts of the polynucleotide chah: the sugar-phosphate backbone, the glycosylic bond between the sugar and the nitrogen base, and certain atoms of the nitrogen base.
The modifications to the native structure of DNA are referred to as lesions. The type of modification created depends on the nature of DNA damaging agents with the resulting lesions possessing inherently mutagenic or lethal properties.
Mutagenic lesions are modifications to the DNA molecule that increase the probability that the replicative polymerases will incorporate the incorrect base during replication. The phenotypic result of a mutagenic lesion may depend on the chrornosomal locus at which it occurred, and on how the base sequence is affected. Ce11 death that results from the presence of certain lesions, is attributed to the ability of lesions to create strand breaks or replication blocks.
Such lethal lesions cause replication blocks due to the altered helical structure
and must be resolved to prevent a lethai consequence (Sagher & Strauss, 1983). Removal of DNA lesions requires recognition by one of the DNA repair
pathways, which proceeds to remove the damage. Therefore, the effect of a
cellular assault by DNA damaging agents depends on the nature of the lesion
created, as well as how the ce11 processes the damage.
The first link behveen DNA damage and pathogenesis came prior to the elucidation of the DNA structure. Observations were made that linked certain
medical pathologies with exposure to specific compounds. Hence, the initial
focus of DNA damage was phenotypic rather than biochemical. Documentation of the pathologies found in soldiers of the First World War correlated their
diseases with exposure to mustard gas (Reviewed in Lawley & Phillips, 1996).
Other early investigators attempted to find compounds that could produce
tumors in test systems. Compounds isolated from tars were responsible for
carcinoma formation in mice; concurrently, studies uncovered the mutagenic
potential of X-rays (reviewed in Lawley & Phillips, 1996). With the technology
at the the, researchers were limited to the isolation of carcinogens, without any
insight into their mode of action. The increasing incidence of cancer and other
genetic diseases in the general population, led epidemiologists to search for
exogenous, synthetic DNA modifying agents that could be responsible for
destabiiizing Our genomes. Exogenous sources of DNA damage are prominent causes of certain types of lesions as exemplified by benzo[a]pyrene in cigarette smoke. However, a substantial amount of DNA damage is also incurred from byproducts of normal metabolic processes (Jaruga & Dizdaroglu, 1996). The DNA damage caused by endogenous agents is as harmful to the genomes of
prokaryotes and eukaryotes as exogenous DNA damaging agents, and is a
constant destructive presence in Our cells.
1.2.2 DNA Damage: Biological Consequences
The characterization of DNA damaging agents is complex due to the
multiple types of DNA modifications one agent cmcause. The pleiotropic
nature of damage created by a certain DNA damaging agent will be manifested
according to: i) the percent of each type of lesion created, ii) location of the
lesion within the genome, and iii) the efficiency of the repair pathways that
recognize the lesions created. Despite the diversity of lesions created by some
DNA damaging agents, the DNA lesion with the greatest biological consequence classifies the agent. Classification of the known DNA damaging agents gives a predicated pattern of genetic modifications which can be used in gene tic s tudies.
1.2.3. Endogenous DNA damage
Despite our preconceived notion that synthetic and environmental agents are the sole cause of DNA darnage leading to carcinogenesis, lesions can arise spontaneousiy from errors in basic cellular processes (le. replication) and from the byproducts of cellular metabolism. The process of DNA replication is required for genome duplication prior to ceIl division and for replacement of
DNA removed duririg excision repair of DNA lesions (Section 1.4.2.).
Replication errors can be attributed to the inherent nature of the DNA polymerases and their limited proofreading ability, and to replication of nitrogen bases that have been modified by DNA damaging agents. Mistakes made by the replication polymerases, such as the misincorporation of a normal base in the daughter strand, is usually removed by the polymerase 533' exonuclease proofreading function or by mismatch repair (Reviewed in Friedberg et RI., 1995). Aside from polymerase errors, a problem arises when there is damage to the pool of dNTP's or to the bases in the DNA template strand. In vitro studies have shown that under physiological conditions the chemical structure of the nitrogen bases is constantly changing, and certain modifications may not be properly recognized by the polymerase (Reviewed in
Friedberg et al., 1995). Depending on the specific modification, the altered base could lead to a mutation or could cause ce11 death upon the next round of replica tion.
The most common type of spontaneous DNA base damage is the generation of abasic (apurinic or apyrirnidinic) sites. Compared to RNA, DNA has an increased resistance to cleavage at the phosphodiester bond, but is more susceptible to cleavage at the glycosylic bonds (Lindahl, 1993). Abasic sites occur when the glycosylic bond is broken, resulting in the loss of the nitrogen base. Purines are released at a higher rate than are pyrimidines, and the structural context (i.e. single-stranded versus double-stranded template) is not a factor (Lindahl, 1993). Abasic sites are believed to occur under normal conditions in the genome of al1 organisms. Their detrimental effect is two-fold: abasic sites are lethal replication blocking lesions (Sahger & Strauss, 1983), and are prone to strand breakage which is also lethal if not repaired (Lindahl, 1993).
Abasic sites can also possess a mutagenic potential; the mutagenic effect is due to error-prone polymerases bypassing the abasic site, incorporating the incorrect base across from the abasic site (Nelson et a!., 1996b).
Tautomeric shifts represent another type of spontaneous genetic modification; they occur when base atoms undergo a spontaneous bonding rearrangement, affecting the normal Watson-Crick base pairing (Friedberg et al.,
1995). The isomers of normal bases that result after the tautomenc shift may forrn hydrogen bonds with incorrect bases. If the base isomer is not removed prior to replication, the incorrect base will be incorporated across from the modified base by the polymerase; if the misincorporated base of the daughter strand is not removed, the lesion is fixed into a mutation.
Loss of the amino group from the nitrogenous base (deamination) occurs spontaneously in the cell, and often changes the coding properties of the base
(Friedberg et al., 1995). It is postulated that deamination occurs by an alkali- catalyzed hydrolysis followed by an acid-catalyzed reaction between water and the protonated base (Lindahl, 1993). Cytosine, adenine, and guanine contain an exocyclic amino group. Deamination of cytosine, resulting in the formation of uracil, is the most common deamination reaction in the ce11 (Lindahl, 1993). The conversion of cytosine to uracil often results in a transversion mutation, since adenine is incorporated opposite uracil (Friedberg et al., 1995). Thymine is the deamination product of 5-methylcytosine, a naturally occuring analogue of cytosine; this reaction results in a transition mutation. The deamination of adenine and guanine results in hypoxanthine and xanthine, respectively.
Hypoxanthine is potentially mutagenic and xanthine arrests DNA replication and is therefore lethal (Friedberg et al., 1995). Reactive oxygen species are created during normal aerobic metabolism, and their cellular consequences are referred to as oxidative damage (Croteau &
Bohr, 1997; Lindahl, 1993). Reactive oxygen species create different types of
DNA lesions with pleiotropic consequences. Of the reactive species created, the hydroxyl and superoxide radicals are primarily responsible for damage to DNA.
To generalize, hydroxyl and superoxide radicals create single and double strand breaks, damage to the deoxyribose moiety, lethal abasic sites and various base modifications (Croteau & Bohr, 1997; Lindahl, 1993). The base modifications caused by these oxygen radicals are varied in their effcct on the cell: i) ring- saturated pyrimidines (predominantly thymine glycols) stall RNA and DNA polymerases, and are consequently non-coding, lethal lesions (Lindahl, 1993); ii)
8-hydroxyguanine (8 oxoG) is the most abundant oxidative lesion and is considered to be a premutagenic lesion due to its ability to mispair with adenine, creating a G+T trawversion mutation (Lindahl, 1993), and iii) the irnidazol ring-opened product of guanine, 2,6-diamino-4-hydroxy-5- formamidopyrimidine (FaPy) is a lethal, replication blocking lesion (Reviewed in Friedberg et OZ., 1995).
The above mentioned genomic events are constantly occurring in living ceils, but the damage they can cause is effectively limited due to the existence of
DNA damage repair pathways. These DNA darnage repair pathways are not only required for spontaneously arising lesions, but also to remove the lesions created by elements in the environment. A brief introduction to the classification of environmental DNA damaging agents and their effect on DNA follows. 1.2.4. Environmental DNA damaging agents
1.2.4.1. Physical Agents
The environment outside a ce11 or organism contains many natural and artificial DNA damaging agents. The most ubiquitous source of environmental
DNA damage is solar ultraviolet radiation (UV). Solar radiation is comprised of
the following wavelengths: W-A (400-320 nm), UB-B (320-290 m),and UV-C
(290-100 nm). UV-C is almost completely absorbed by the ozone layer; the wavelengths of biological significance are UV-A and UV-B, and will be referred
to collectively as UV (Friedberg et al., 1995). UV creates different types of lesions depending on the DNA substrate. Cyclobutane pyrimidine dimers occur when bonds are formed between adjacent pyrimidine residues. There are four stable configurations of dimers: cis-syn, cis-anti, tram-syn, and trans-anti; al1 four configurations cause DNA helix distortions that are lethal replication blocks
(Berends, 1961). Gibbs et 01. (1993) have shown that there is a difference between how cells repair the cis-syn pyrimidine dimers and the trans-syn pyrimidine dimers, indicating that the type of configuration is biologically significant.
Another major product of WV irradiated DNA are 6-4 photoproducts. These dimers are a noncyclobutane type of di-pyimidine photoproduct in which there is bond formation between the C6 and C4 of adjacent pyrimidine bases
(Friedberg et al., 1995). The two prominent forms occur between adjacent T-C and C-C bases, causing lethal replication blocking distortions to the helix
(Rabkin, et al., 1983; Chan et al., 1985; Larson & Strauss, 1987). Less prominent
W lesions include different forms of base darnage, DNA-protein crosslinks, and DNA strand breaks (Friedberg et al., 1995). Ionizing radiations (IR) from natural and artificial sources create a wide range of DNA lesions. Production of DNA lesions by IR can be direct or indirect but often have the same cellular consequences. The energy of ionizing radiations can catalyze reactions by exciting the electrons of cellular molecules.
Direct damage by IR results when electrons are abstracted from part of the DNA molecule. Additionally, the abstraction of electrons from water creates many
DNA damaging free radicals, including 'OH, 0,- and H,O, (Friedberg et al.,
1995). Indirect damage occurs through the creation of free radicals resulting from ionization of other cellular molecules. The free radicals produced by IR have the same cellular consequences as the DNA damaging reactive oxygen species created during normal cellular metabolism. Similar to the lesions created by endogenous oxidative damage, ionizing radiation can cause base damage, as well as single- and double-strand breaks (Croteau & Bohr, 1997).
1.2.4.2. Chernical Agents
DNA alkylating agents are electrophilic compounds that donate alkyl groups to different nucleophilic residues on the nitrogen bases of DNA (Reviewed in Lee et al., 1992). Certain residues on the nitrogen bases are more susceptible than others, depending on their nucleophilic strengths; N~ and N' of purines are strong nucleophiles. 060f guanine and O2and Olof thymine are moderate nucleophiles (Lee et al., 1992). Reactivity of alkylating agents with
DNA is summarized by their Swain-Scott substrate constant (s). Alkylating agents with low s values have reduced reaction selectivity and are referred to as
SNI cornpounds. Alkylating agents with s values approaching 1 are referred to as SN2compounds. Ethylating agents (donate ethyl- groups to DNA base) tend to have lower s values than do methylating agents (Lee et al., 1992). The classification of SNIand SN2alkylating agents does not specify the product, but rather helps define the distribution of lesions created. The type of lesion produced is of importance due to the cellular consequences; certain lesions created by alkylating agents are lethal and some are mutagenic. Despite the varied nature of the lesions created by alkylating agents, there are some S,1 and
SN2alkylating agents that are well characterized in the patterns of lesions they create, allowing them to be used for DNA damage studies. Methyl methanesulfonate (MMS) and N-methyl N-nitro-N nitrosoguanidine (MNNG) are two well characterized chemicals used in DNA damage studies that represent SNIand SN2alkylating agents, respectively. Major lesions of biological significance created by MMS are 3-methyladenine (3MeA) and 7-methylguanine
(7MeG); 3MeA stalls replication when encountered during S-phase of the cell cycle and is therefore lethal. Both 3MeA and 7MeG form secondary lethal lesions because they are easily converted to abasic sites due to the weakened glycosylic bond, which then may lead to strand breakage. MNNG produces many different alkylation products, but also creates a significant level
06rnethylguanine (06MeG). 06MeG base pairs with thymine, and is therefore a mutagenic lesion.
Cross-linking agents are bifunctional chemicals that can bind simultaneously to two different sites on the DNA molecule. If bond formation occurs between sites on the same DNA strand, these lesions are intra-strand cross-links; bond formation between two separate strands produces inter-strand cross-links. Intra-strand and inter-strand cross-link lesions are both lethal due to their ability to stall replication and prevent strand-separation, respectively
(Reviewed in Friedberg el al., 1995). Nihogen mustard, infamous for its toxic effects on the soldiers of the First World War, is a cross-linking agent that creates both inter- and intra-strand cross-links (Reviewed in Lawley & Phillips,
1996). Cisplatin is another weli known cross-linking agent used in chemotherapy. Cisplatin also creates both inter- and intra-strand cross-links, with predominantly more intra-strand cross-links (Reviewed in Lawley & Phillips, 1996).
Many more classifications of DNA damaging agents exist. Intercala ting drugs incorporate into the DNA molecule and sequester the same amount of space as a normal base pair. Due to their physical property, intercalating agents such as acridine orange cause frameshift mutations (Reviewed in Friedberg et nl., 1995). There are exogenous agents that can penetrate cells and create reactive oxygen species, causing the sarne genomic consequences as the endogenous byproducts of aerobic metabolism (Reviewed in Friedberg et ni.,
1995). The categories of DNA damaging agents is not exhausted. The above mentioned sources of DNA damage represents the most common types used in laboratories, due to their biological significance or the well established mechanism of damage.
1.3. How the Ce11 Copes with DNA damage
Despite the fact that DNA damage occurs as a natural consequence of both ceIlular metabolism and of unavoidable elements in the environment, it is not favorable to allow the DNA damage to persist. We are more aware of the presence of exogenous agents that may elevate the level of DNA lesions, and threaten the integrity of our genomes. Our exposure to DNA damaging agents cannot be completely controlled, amplifying the importance of cellular DNA damage control. Regardless of the source of damage, most organisms respond to DNA damage or to blocked DNA replication with three modes of defence: i) arrest of ce11 cycle progression, ii) inhibition of late-initiated replication, and iii) upregulation of DNA damage repair genes with subsequent repair of the damage (Weinert, 1998). Coordination of ce11 cycle arrest and DNA damage repair is critical, since successful repair relies on the activity of both mechanisms.
1.3.1. Ce11 Cycle Controls
The ce11 cycle is the progression of a ce11 from one mitotic generation to the next; each generation is characterized by four distinct phases: G1, S, G2 and
M. Collectively, G1, S and G2 have been known as interphase, the phase bridging successive mitoses (Nasmyth, 1996). S phase is the stage of genorne duplication that is preceded and followed by two gap phases, G1 and G2, respectively. Ce11 division, the mitosis phase (M phase), occurs after G2, and once completed produces two G1 phase daughter cells. The transitions between the phases of the ce11 cycle are strictly controlled. Progression of each phase is not initiated until the events that characterize the previous phase have been completed (Reviewed in Nasmyth, 1996). The sequential activation and inactivation of enzymes required for ce11 cycle progression events prevents ce11 division with irnproperly replicated DNA, and allows the ce11 time to complete the task of replication properly. In eukaryotes, and possibly in prokaryotes as well, there exist surveillance systems that monitor the state of the genome and can prevent ce11 cycle progression if any delays occur, or if DNA damage is sustained (Hartwell
& Weinert, 1989). Cell cycle progress surveillance and ce11 cycle arrest is executed by checkpoint enzymes. Checkpoint proteins recognize DNA damage or premature ce11 cycle progression, and stall further progression until the genomic and cellular environment is secure and ready to precede to the next stage. Checkpoints are therefore crucial to the genomic stability of the cell. In S. cerevisine some enzymes are involved in al1 ce11 cycle checkpoint signal cascades
(Mecl, Rad53), and other enzymes are specifically required by one or two of the three characteristic checkpoints. The DNA-damage dependent ce11 cycle arrest points occur at the G1-S transition, the S-M transition and the G2-M transition.
Once the checkpoint machinery detects genetic abnormalities within the context of the specific phase of the ce11 cycle, the signal transduction cascade stops further cell cycle progression. Important enzymes involved in "sensing" DNA darnage and signalling ce11 cycle arrest include Rad9, Radl7, Rad24, Rad53,
Mecl, Mec3, Pole, and Pdsl (Reviewed in Paulovich et al., 1997b). DNA repair pathways in cooperation with these ce11 cycle controls prevent possible genomic alterations ranging from minor base substitutions to gross rearrangements.
The effectiveness of DNA repair is dependent on ce11 cycle controls, and vice versa. DNA repair requires a delay in the ce11 cycle to properly remove the
DNA lesion. Without allowing tirne for the various repair enzymes to recognize and repair DNA damage, the cells could endure the effects of the lesions
(mutagenesis or ce11 death) or could allow progression with harmful repair intermediates (Weinert, 1998; Paulovich et al., 199%). Ce11 cycle arrest without DNA repair is useless; the lesions persist independently of how long the ce11 cycle is stalled. In S. cerevisiae the length of arrest with unrepairabie damage is ahost 10 hours (Sandel & Zakian, 1993; Toczyski et al., 1997); after this period of time the ce11 cycle progresses with varied consequences depending on the current state of the ce11 cycle and the type of damage present. Different types of DNA damage can occur during the different phases, and this can depend on the metabolic events occuring during the particular phase and the duration of the phase. In S. cermisiae G1 is the longest phase in the ce11 cycle; this metabolically active time gives rise to high levels of reactive oxygen species known to damage DNA (Nasmyth, 1996). Any repair that is to occur in G1 must be done in the context of a haploid genome, making some forms of recombination repair lethal to the cell. Damage incurred during the S phase of the ce11 cycle presents the ce11 with an extra complication; in the presence of damage, replication fork progression without ce11 cycle arrest may lead to lethal destabilized replication forks, or a mutational event. Most DNA damage repair pathways ca~otrepair damaged DNA in the context of a replication fork. Unlike damage in G1, damage in the G2 phase of the ce11 cycle occurs to a diploid substrate. The phase of the ce11 cycle in which damage occurs can affect how lesions are repaired, influenced by the cellular substrates present. Once the ce11 cycle checkpoint proteins arrest the ce11 cycle, the DNA repair enzymes are needed to repair the damage; this ensures that no damage will be inappropriately repaired or ignored prior to ce11 cycle progression. 1.3.2. DNA Damage Repair
To keep the level of genornic modification as low as possible, al1 living cells are equipped with various DNA repair pathways. DNA repair pathways consist of enzymes that work in concert to identify and remove damaged DNA.
The number of enzymes required in a pathway may relate to the spectrum of lesions recognized, or to complexity in the repair mechanism. The specificity of each repair pathway is evident when observing the dmg sensitivity of mutants in the various pathways. As previously mentioned, the importance of DNA repair is well illustrated in the following two observations: genes involved in the various DNA repair pathways are highly cowerved throughout al1 species and DNA repair deficiences have been directly linked to genetic diseases. The evolutionary conservation of DNA repair pathways enables researchers to study complex pathways in uni-cellular systems, reducing the difficulty that would be encountered in multi-cellular organism.
1.4. DNA Repair Pathways
1.4.1. Direct Reversa1 of DNA Damage
There are DNA repair pathways that function to remove the lesion without additional alteration to the native DNA molecule. The types of lesions that are repaired by this category of pathway are very specific due to the limited lesion recognition capabilities of their enzymes. The two pathways discussed are important in the removal of DNA modifications caused by UV and by alkylating agents. A W pyrimidine dimer is lethal to the ce11 if it is not removed before being encountered by the replication machinery (Berends et al., 1961). The reversa1 mechanism for pyrimidine dimer removal is photoreactivation. The process of photoreactivation exists in E. coli and S. cerevisiar ,but not in humans.
The photolyase of Snccharonzyces cerevisiae is encoded by the PHRl gene; the enzyme recognizes cyclobutane pyrimidine dimers and repairs them by deaving in a light-dependent manner the hydrogen bonds that join adjacent pyrimidine bases (Friedberg et RI., 1995). The photolyase cleaves the hydrogen bonds of the dimer by flipping the damaged bases out of the helix, and into the active site of the enzyme. E. coli cells have a similar photolyase enzyme (sharing 36% identity) also involved in light-dependent cyclobutane pyrimidine dimer removal.
Another example of DNA damage repair utilizing a reversa1 mechanism involves a subset of enzymes that repair alkylation damage. Certain lesions formed by alkylating agents are removed by the alkyltransferase family.
Alkyltransferases remove the modifying alkyl group from the DNA molecules by accepting the alkyl group at a cysteine residue in their active site (Reviewed in Pegg & Byers, 1992). Once the modifying group is transferred to the alkyltransferase, it cannot be removed and the protein is no longer hnctional.
This method of repair is not energy efficient, but is important for removhg mutagenic lesions without any modification to the native DNA molecule.
Alkyltransferases have been isolated in bacteria, yeast and mammalian cells
(Reviewed in Pegg & Byers, 1992). The o6Methyltransferase (Mgtl) of S. cereuisiae acts specifically on 06-methylguanine and O4-rnethylthymine lesions
(Sassanfar et al., 1991). Removal of the methyl groups from guanine and thymine has no effect on the native structure of the helix. Disruption of the MGTl open reading frame (ORF) confers a MMS and MNNG sensitivity, the dmgs responsible for creating O6-MeG and O4-MeT.
The presence of enzymes like photolyase and Mgtl represent cellular decadence, and should contradict the theories behind evolution. Both of these enzymes in S. cerevisiae are not recycled, and only recognize a minute subset of
DNA lesions. The evolutionary persistence of enzymes that repair by reversing the damage incurred by DNA is indicative of their importance; both photolyase and Mgtl recognize and repair biologically significant lesions in a time-sensitive fashion. Avoiding the mutagenicity of 06MeG and the toxicity of pyrimidine dimers is worth the extra energy sequestered from the ce11 for these enzymes.
1.4.2. Excision of DNA Damage
There are two forms of excision repair in both prokaryotes and eukaryotes, base excision repair and nucleotide excision repair. Both pathways remove lesions by excising the damaged sequence out of the DNA, with subsequent replication of a new undamaged segment. Each pathway is responsible for the rernoval of specific lesions, with little overlap in substrate recognition; mutant cells with defects in both excision pathways have sensitivities to a wide array of DNA damaging agents. The base excison repair pathway recognizes and repairs lesions that are relatively small modifications of the DNA. Nucleotide excision repair predominately recognizes lesions that cause helical distortions. Both pathways are necessary for the proper repair of lethal and mutagenic genomic alterations. 1.4.2.1. Base Excision Repair Al1 base excision repair (BER) is initiated through the recognition and removal of damaged bases by DNA glycosylase enzymes. Glycosylases bind to
DNA and mediate the cleavage of the damaged bases from the sugar backbone
(Reviewed in Seeberg et al., 1995). DNA glycosylases from E. coli ,S. cereuisine, and mammalian cells have been purified. The types of lesions recognized by
BER is dependent on the specificity of each glycosylase in the respective organism. At least seven glycosylases have been identified in E. coii :tngl encodes a glycosylase that recognizes 3MeA, uqencoded glycosylase recognizes misincorporated uracil, the AlkA 3MeA glycosylase recognizes
3MeA, 7MeG, 2MeC, 5-formyluracil and other minor DNA damaged bases, MutM recognizes FaPy, 8-oxoG, and 5-hydroxycytosine, endonuclease III and N recognize urea, thymine glycols and 5-hydroxycytosine, and the A-G mismatch DNA glycosylase recognizes mismatched adenines and 5-hydroxycytosine
(Seeberg et al., 1995). The seven glycosylases recognize a wide range of DNA lesions, but some do share the substrates they recognize. The substrate recognition overlap eliminates the potential for the ce11 to neglect the repair of biologically significant lesions.
S. cereuisine has at least five DNA glycosylases; the uracil glycosylase recognizes uracil, the Magl glycosylase has a similar substrate preference to the E. coli 3MeA DNA glycosylase (AlkA), Oggl recognizes FaPy and 8-oxoG, and the Ntgl and Ntg2 glycosylases recognize a number of pyrimidine-derived lesions ( 5-hydroxyuracil,5-hydroxycytoshe,and thymine glycol), as well as
FaPy and 8-oxoG (Seeberg et al., 1995; Senturker et al., 1998). As with the E. coli glycosylases, there is overlap in substrate specificity in the S. cerevisiae glycosylases. The above mentioned lesions recognized by these DNA glycosylases are of biologicaI sigmficance since both mutagenic and toxic lesiow constitute the substrates of BER.
The resultant abasic site created following cleavage by the DNA glycosylases must be processed to prevent the persistence of this lethal intermediate. The abasic site is processed by one of two enzymes: i) some glycosylases possess an associated lyase activity, which cleaves the sugar- phosphate backbone 3' to the abasic site, or ii) the apurinic/apyrimidinic endonucleases (AP endonucleases encoded by APNl and APN2) cleave the sugar-phosphate backbone 5' to the abasic site (Seeberg et OZ., 1995; Friedberg et al., 1995). Cleavage by an AP endonuclease leaves a structure that must be further processed by a 5' phosphodiesterase. Once the strand break has been created, a short patch or long patch BER mechanism processes the structure. In short patch repair, the single nucleotide gap is filled in by a DNA polymerase and the strand break is subsequently ligated. In long patch BER, DNA polvrnerase fills in the missing base and continues to polymerize a few nucleotides downstream, displacing the existing strand. The resulting oligonucleotide flap is cleaved off by the Fenl /Rad27 enzyme, and the strand break is ligated together (Reviewed in Wilson & Thompson, 1997).
BER is an important pathway for the avoidance of mutagenic and lethal genetic events; its importance is observed in the phenotypes of BER mutants.
Deletion of the various DNA glycosylase genes increases the cellular sensitivity to alkylating agents, oxidatizing agents and to any endogenous damage
(Seeberg et al., 1995). Mutants defective in the AP endonucleases have an increased spontaneous mutation rate and sensi tivi ties to alkylating and oxidatizing agents (Xiao & Samson, 1993). ln AP endonuclease mutants the abasic sites formed by the action of glycosylases are not resolved, leaving the lethal abasic sites to be processed by other DNA damage repair pathways.
1.4.2.2. Nucleo tide excision repair
The nucleotide excision repair (NER) pathway is primarily responsible for the repair of DNA darnage created by UV, W-mimetic agents, and cross- linking agents. The commonality between these DNA damaging agents is that they al1 cause helix-distorting lethal lesions. NER is an important pathway in preventing DNA damage-induced ceil death. The NER pathway recognizes DNA damage and repairs the lesion by excising approximately 30 nucleotides from the damaged DNA (the strand containing the damage), with subsequent resynthesis to replace the excised segment. There are two classes of NER: repair associated with transcribed DNA
(transcription-coupled repair; TCR-NERI and repair associa ted with untranscribed DNA (global genome repair; GG-NER) (deLaat et RI., 1999). Both classes of NER require the following activities for proper repair: lesion recognition, strand separation between damaged and undamaged strands, excision of the damaged sequence and polymerization of DNA to replace the excised sequence.
Lesion recognition is mediated differently in TCR- versus GG-NER. In
TCR-NER, the enzymes involved in lesion recognition are unknown. It is hypothesized that the signal for the repair of DNA damage is simply the blocking of RNA polymerase II by these lesiow, which then triggers the NER cascade. In GG-NER, lesion detection is performed by the Rad4-Rad23 (S. cerevisiae) or XPC-hHR23B (hurnan) complexes (Araujo & Wood, 1999; deLaat et al., 1999). Once darnage has been recognized, the strand containing the damage must be separated from its undamaged complementary strand, as a stabilized open helical structure. Strand separation is mediated by the TFHH holoenzyme
(S. cerevisiae and humans), which contains 5' and 3' helicases. TFZIH has a dual cellular role: it is a necessary component of RNA polymerase II mediated transcription and is required foi al1 NER. Once the strands aeseparated, other factors help stabilize the repair bubble. At the bubble, the ERCCI-XPF (human) or RadlO-Rad1 (S. cerevisiae) and XPG (human) or Rad2 (S. cereuisirre) endonucleases cleave the damaged DNA strand 5' and 3' to the damage, respectively (deLaat et al., 1999). The excision created by ERCC1-XPF/RadlO-
Rad1 allows for DNA synthesis in the resulting gap from the -OH group at the 3' terminus. Replication stops at the 3' cleavage site (deLaat et al., 1999).
The importance of the NER pathway is demonstrated in the correlation made between human xeroderma pigmentosum and defects in enzymes of NER
(Cleaver, 1968). People with xeroderma pigmentosum are extremely photosensitive and have an increased rate of developing skin cancer at a young age; neurological abnormalities are also associated with xeroderma pigmentosum (Friedberg et al., 1995). Patients with xeroderma pigmentosum fa11 under one of eight complementation groups. Seven of the complementation groups define a defective NER gene and denoted as XP-A, -B, -C, -D, -E, -F or -G (Friedberg et al., 1995). The final complementation group, XP-V (variant), is the only one which does not possess a defect in NER (Cleaver, 1972). 1.4.3. Recombination Repair The above mentioned repair pathways employ efficient mechanisms to repair the damaged base moieties of DNA by excising the damage out of the helix. These repair pathways are of little benefit when the type of damage incurred is single or double strand breaks. Clastogenic agents define DNA damaging agents that create single and double strand breaks (DSBs), including y-rays and bleomycin. Conversion of a single strand break (SSB) to a DSB (via replication of a nicked template), or persistence of a DSB, eventually leads to ce11 death (Obe et al., 1992). Many enzymes involved in normal homologous recombination are used in S. cerevisiae for the repair of DSBs; higher eukaryotes possess the homologous enzymes, but prefer another mechanism for DSB repair known as non-homologous end joining.
Homologous recombination involves the genetic exchange of homologous sequences between sister chromatids or homologous chromosomes, resulting in transient base pairing or permanent sequence exchange. The initiating event in recombination is the generation of a free DNA end; the linear DNA end invades and base pairs with the homologous complementary strand of a duplex DNA molecule. The second strand of the
DNA duplex is displaced by the invading strand, creating a D-loop structure.
The invading strand can then act as a primer for replication, with hrther displacement of the original strand. Resolution of the strand invasion event can result in the dissociation of the invading strand from the D-loop, with subsequent religation to its original complementary strand. Permanent sequence exchange can occur if the Holliday junction is resolved by endonucleases (Reviewed in Flores-Rozas & Kolodner, 2000). Holliday junctions represent the cross-over point of DNA hybrid duplexes formed during homologous recombination (Holliday, 1964). When a double strand break occurs, the ce11 must ensure that no sequence between the two free ends is missing. The best way the ce11 can assess whether any intervening sequence is missing is to compare the damaged sequence with the sequence of a sister chromatid or a homologous chromosome. In S. cereuisiae, the enzymes required for homologous recombination are used to facilitate invasion of the damaged strand into the duplex of a homologous chromosome. The enzymes required for homologous recombination are encoded by RAD51, -52, -54, -55 and -57 (Reviewed in Shinohara & Ogawa,
1995). The initiating event for recombination is a free DNA end, which is produced by the damage. Rad51 (homologue to E. coli RecA) is required for the base pairing of a single stranded DNA molecule with the homologous DNA duplex molecule (creating heteroduplex DNA ); its activity is stimulated by the
Rad54 ATPase (Petukhova et al., 1998). Rad52, Rad55 and Rad57 are believed to enhance the ability of Rad51-Rad54 to bind to areas of single stranded DNA, which is necessary for heteroduplex formation (Sung, 1997). This process ensures that any possible missing sequence is replaced by replicating from the homologous duplex molecule sequence adjacent to the cleavage (Shinohara & Ogawa, 19%).
Mutants defective in any genes of the RAD52 family are sensitive to clastogenic agents (X-rays, bleomycin, MMS and similar DNA damaging agents that create DSBs). Cells treated with these agents have elevated recombination levels, providing additional evidence that recombination repair is respowible for DSB repair (Rodarte-Ramon & Mortimer, 1972). As can be expected, recombination repair is most effective during the G2 phase of the ce11 cycle or in a diploid genome, since in both instances the homologous sequence is available for recombina tion.
1.5. DNA Damage Tolerance
The lesions typically repaired by NER and BER are often lethal if encountered by DNA polymerases. If any of the enzymes in either pathway are saturated or unable to repair such lesions, cell death will result. Similarily, if the ce11 is bombarded by agents that create replication-blocking lesions during the S phase of the ce11 cycle, it is unlikely that the lesions will be repaired before stalling a replication fork. To prevent ce11 death in such circumstances, al1 cells contain DNA damage tolerance pathways. DNA damage tolerance may also serve to protect a ce11 against any unforeseen DNA lesion that may not be recognized by the known repair pathways (Walker, 1995). The "better safe than sorry" philosophy may be important in an environment with new chemical threats, since there may arise lesions that BER and NER camot repair.
The phenornenon of DNA damage tolerance has been observed in both prokaryotes and eukaryotes. In both cases, excision-repair (NER) deficient cells produce low molecular weight DNA replication products after UV-irradiation
(Rupp et al., 1971; Prakash, 1981; DiCaprio & Cox, 1981). After a short incubation period the fragmented genomic DNA converts to larger molecular weight species, similar to the DNA of unirradiated controls (Rupp et al., 1971;
Prakash, 1981; DiCaprio & Cox, 1981). The low molecular weight DNA was detected using alkaline sucrose gradients, and is assumed to arise from stalled replication forks creating areas of single stranded gaps (Rupp et al., 1971; Prakaçh, 1981; DiCaprio & Cox, 1981). DNA damage tolerance is the DNA replication-dependent process that resolves the regions of single shanded DNA; it recognizes and processes damage in the context of a stalled replication fork
(Kadyk & Hartwell, 1993).These tolerance pathways do not remove the damage within the replication fork; instead they allow for resumption of replication while the lesion persists. The dimers responsible for stalling replication are not removed from the DNA, and persist even after the gaps are resolved (Ganesan,
1974; Sarasin & Hanawalt, 1980). Conversely, in cells defective in DNA damage tolerance, the damage can be proficiently excised by NER, but the cells are not protected from damage incurred during DNA replication; the regions of single stranded DNA are not resolved (Prakash, 1981). A subset of genes under the control of the SOS DNA damage response regulon are responsible for DNA damage tolerance in E. coli . The mechanisms of these gene products are well characterized, and have been used as a mode1 for mechanistic comparisons in other organisms. In contrast, the postreplication repair pathway (PRR) in S. cereuisine is poorly understood.
1.5.1. SOS Response in E. coli
ïhe DNA damage tolerance activity in E. coli is encoded by a set of genes that are part of the SOS regulon. These genes are induced in response to regions of single stranded genomic DNA (Chaudhury & Smith, 1985). Single stranded
DNA can be created directly by aberrant replication, or by DNA damage that leads to discontinuous replication. The two SOS-dependent mechanisms of
DNA damage tolerance are the translesional replicative bypass (mutagenic pathway), and recombinational bypass. Recombination b ypass is the predominant form of gap repair in E. coli . The RecA protein is essential for both
mechanisms of DNA damage tolerance, since it is responsible for activation of
the SOS response and is necessary for al1 recombinational events. RecA binds to single stranded DNA, promoting base pairing between the single stranded
DNA-RecA nucleoprotein filament and a homologous DNA duplex. RecA provides E. coli with the analogous activity of the Rad51 protein in S. cerevisine
(Roca & Cox, 1997).
RecA-dependent recombination involved in DNA repair can be processed through one of two parallel pathways: the RecBCD pathway and the
RecFOR pathway (Cox, 1997). The recombination rnediated by the RecBCD enzymes appears to be important in the repair of DSBs, lesions created by ionizing radiation (Michel et al., 1997). The RecFOR pathway is associated with repair of single stranded gaps (Lanzov et nl., 1991) and in the repair of interrupted replication forks (Courcelle et al., 1997; Courcelle& Hanawalt, 1999).
When replication arrests, the resultant single shanded DNA is bound by
RecA. Once RecA binds to the section of ssDNA, it becomes activated
(Sassanfar & Roberts, 1990). Activated RecA (RecA*)not only facilitates homologo~srecombination-mediated replication restart and gap filling, but also initiates the SOS response. Activated RecA promotes the self-cleavage of the
LexA protein, the enzyme responsible for traiiscriptional repression of the genes within the SOS regulon. RecA* is not a protease, but stimulates the auto- cleavage of LexA (Reviewed in Friedberg et al., 1995). LexA cleavage removes the repressor from the promoters of a wide variety of genes, including recA and the genes of the mutagenic DNA damage tolerance response (umuC, nmzrD, and dinB) (Reviewed in Friedberg et al., 1995). RecA" in cooperation with RecF, RecO and RecR (RecFOR) promote the re-initiation of the stalled replication fork by obtaining the coding information for the missing sequence from the homologous duplex. The RecFOR enzymes are involved in the stabilization and re-initiation of the stalled replication fork
(Reviewed in Kogoma, 1997). Other enzymes involved in this form of gap filling include the prirnosome complex proteins, whose function is confined to recombination-dependent replication restart (Zavitz & Marians, 1991).
The mechanisms of recombination-dependent DNA damage tolerance are fairly well understood. More recently the second pathway of DNA damage tolerance in E. coli was elucidated biochemically. This translesional replicative bypass pathway involves error-prone polymerases to resolve regtons of single stranded DNA. umuC, iimuD and dinB are the genes that are derepressed as part of the SOS -induced mutagenic respowe. These genes encode polymerases capable of synthesizing DNA past sites of DNA damage. DinB (polymerase IV) is a non-essential polymerase that appears to be inherently mutagenic, creating mutations in DNA that have not been exposed to DNA damaging agents.
Mismatch mutations be tween primers and templates, and replica tion pause sites can be replicated by DinB (Wagner el al., 1999). DinB-dependent replication often leads to -1 frameshift mutations. UmuC and UmuD form a heterodimer that, Ui cooperation with RecA* and single stranded binding protein, replicates past regions of DNA damage (Tang, 1999; Bacher Reuven et ni., 1999; Maor-
Shoshani et al., 2000).
The mutagenic pathways of the SOS response handle DNA damage differently from the recombination-dependent DNA tolerance pathways. In the mutagenic response, the stalled replication fork is reinitiated by these non- essential polyrnerases, using the damaged strand as a template. Quite often the wrong nucleotide is incorporated across from the lesion, explaining why this
pathway of the SOS response is mutagenic. In an in vitro system, UmuCD
appear to have a preference for dAMP incorporation (69%) opposite an abasic
site (Bacher Reuven et al., 1999). As one cm imagine, the mutagenic nature of
replication restart is not the preferred tolerance mechanism, since mutagenesis is
associated with genomic instability. For this reason, the mutagenic response in
E. coli is tightly regulated at two levels; it is controlled at a transcriptional level
and at a post-translational level.
Transcriptionally, the genes of the SOS regulon are tightly repressed by
the binding of LexA to their operator sequences. Transcriptional derepression of these genes requires RecA activation, which is contingent upon DNA damage. Once the SOS response has been initiated, LexA is removed from the operators of the SOS regulon genes. Within the genes of the SOS response there are differential levels of derepression upon SOS activation, dependent on the strength of individual operators, orientation of the operator in relation to the promoter, pornoter strength and the presence of other promoters (Friedberg et al., 1995). The operator of unrz~CDhas a higher binding affinity for LexA and
thus a higher level of repression than most of the other SOS genes (Peterson &
Mount, 1987; Schnarr et al., 1991).
Once transcription of the urntiCD operon has occurred, the two
transcripts are translated. For the mutagenic polymerase, UmuC must form a heterodimer with a cleaved form of UmuD (UmuD'). If UmuC complexes with uncleaved UmuD, the heterodimer is unable to act as a polymerase (reviewed in
Friedberg et nl., 1995). UmuD cleavage by LexA is promoted by RecA*, resulting in the large carboxyl terminal domain (UmuD') binding with UmuC
(Burckhardt et al., 1988). This process is important in controllhg elevated levels of UmuC and UmuD that may persist after the detrimental DNA structure has been resolved. Once the DNA damage has been processed, the level of RecA* declines so persistent UmuC and UmuD molecules will not receive the proteolytic signals that are necessary for activation.
The SOS regulon is an elegant example of how the ce11 can regulate DNA damage tolerance. After sustaining moderate levels of DNA damage, the ce11 relies upon error-free mechanisms of DNA damage tolerance; under massive genomic stress, a further tolerance pathway is induced (UmuCD' and DinB mutagenesis) that prevents ce11 death and may offer an option for environmental adaptation in unstable conditions.
1.5.2. Postreplication repair in Eukaryotes
It has been shown in S. cerevisiae, as described in E. coli , that after UV- irradiation there are discontinuities in DNA replication, creating regions of single stranded gaps (Prakash, 1981; di Caprio & Cox, 1981). As observed in E. coli ,these replication-dependent single stranded gaps are resolved by the DNA damage tolerance pathway. To identify an analogous pathway in S. cereuisine, researchers used alkaline sucrose gradients to examine the ability of various mutants to resolve the single stranded gaps created after UV-irradiation
(Prakash, 1981; di Caprio & Cox, 1981). The alkaline conditions in an alkaline sucrose gradient allowed for the detection of single stranded breaks and single stranded gaps produced from incomplete replication, in addition to double strand breaks. The mutants examined were selected based on their extreme sensitivity to a broad spectrum of DNA damaging agents, their independence from other known DNA repair pathways or their analogous activities to known
SOS enzymes (Prakash, 1981; di Caprio & Cox, 1981). These studies determined that postreplication repair in S. cerevisiae is dependent upon the RAD6, RAD18 and RAD52 genes.
It is currently unknown what occurs upon replication stalling. 1s the helicase or polymerase blocked when a lesion is encountered? 1s the polymerase removed from the replication bubble, or is its hction modified to cope with the damage? It is also not clear what mechanism is responsible for gap filling: a translesional bypass mode of replication or a recombination-mediated template switching event that allows the ce11 to replicate past the damage. A discussion of the members of PRR in S. cermisine may help elucidate these questions.
1.6. Postreplication Repair In S~ccharotnycescerevisiae
1.6.1. Recombination repair in DNA damage tolerance
It was previously discussed how the RAD52 epistasis group is involved in recombination (mating type switching and meiotic recombination) and in the repair of DNA DSBs. E. coli preferentially employs recombination-dependent mechanisms to resolve replication blocks, but the role of recombination in eukaryotic PRR is unknown. It is accepted that homologous recombination may serve as a PRR-independent mechanism for resolution of stalled replication forks in a diploid organism or a haploid organism in G2 of the ce11 cycle
(Swanson et al., 1999). The traditional homologous recombination pathway and
PRR pathway are viewed as two mutually exclusive mechanisms. Arguments supporthg and disputing the involvement of recombination in PRR demand more relevant data. Eukaryotes possess a mutagenic DNA damage tolerance pathway (discussed in section 1.6.3.4.2) analogous to the UmuCD' and DinB pathway in E. coli ,but the pathway's role in tolerance appears to be secondary to an ill-defined error-free pathway. Establishing the mechanism of this error- free PRR pathway involves considering a role for homologous recombination.
There are four major arguments for recombination-mediated replication res tart/ gap-filling in eukaryotes. Firs tly, recombination based replica tion restart represenh the error-free mode of DNA damage tolerance in E. coli. Secondly, supporting evidence in eukaryotes includes a documented elevation in recombination in UV-irradiated cells (Snow, 1968), and in the visual observation of DNA damage-induced Holliday structures (Higgins et al., 1976).
Thirdly, the initial survey of PRR mutants by Prakash (1981), indicates that in addition to the rad6 and rndl8 mutants, the rad52 mutant is also defective in PRR activity. The NER- rad52 cells showed a PRR defect upon UV irradiation (Le. an inability to resolve the single stranded DNA alter UV irradiation) (Prakash,
1981). Finally, NER Sncchnromyces cerevisiae cells tolerate replication blocking lesions by utilizing a sister chromatid exchange (SCE) dependent process
(Paulovich et al., 1998), requiring a Holliday junction intermediate formed in a
RA052 -dependent manner (Zou & Rothstein, 1997).
There is little dispute when people define the RAD6 epistasis group as the PRR pathway in S. cerevisine. Rad6 and Rad18 are necessary for error-free and error-prone PRR. The error-prone gap-filling mechanisms are well defined and are discussed in section 1.6.2.3.. However, predicted mechanisrns that would be responsible for the resolution of the stalled replication fork via the error-free pathway are not exciusively fulfilled by Rad6 or Radl8. New candidates for error-free PRR are assessed by their genetic relationship with
Rad6 or Radl8. Interpretation of genetic relationships led to a dispute over whether Rad52 was in fact involved in error-free PRR. Resnick et al. (1981) examined the level of dimer exchange in post-irradiated S. cereuisiae NER cells; the purpose was tu assess whether the sequence containing the pyrimidine dimers was exchanged with the newly synthesized strand, as seen in E. coli (Ganesan, 1974). Since dimer-dependent strand exchange was not observed, it was concluded that a recombination based mechanism was not involved in PRR
(Resnick et al., 1981). Mutants isolated as suppressors of rad6 and rad18 W sensitivities were al1 allelic to the SRS2 gene, which encodes a 3' to 5' heiicase
(Lawrence & Christensen, 1979; Schiestl et al., 1990; Aboussekhra et al., 1989). It is postulated that the Ss2 helicase is required for Rad6-dependent PRR, preventing the activity of other tolerance mechanisrns. Interestingly, the removal of SRS2 in a rad6or rad18 background will only suppress cellular W sensitivity if there is a functional recombination repair pathway (Schiestl et al.,
1990). These data suggest a distinction between PRR and recombination pathways, rather than a cooperative mechanism.
Dissociating recombinational processes from PRR may be presumptive based on the evidence provided above (ie. no observed dimer exchange and the suppression of rad6 and rad18 by recombination repair in an srs2 background).
Models developed to explain resolution of a collapsed replication forks in E. coZi envisage a cooperative recombination-replication based mechanism to reestablish the replication fork. Sister chromatid exchange is one possible process that could bypass a damaged template (Fig. 1A). Another mode1 used
to explain resolution of the replication fork involves template switching with
the interrupted strand using complementary nascent DNA as the template (Fig. 1B). In both models once the lesion has been bypassed another template switching event will reestablish the correct structure, and both models would promote lesion bypass without dimer exchange between strands. New interpretations of old data, combined with more recent findings, may help solve whether the enzymes of recombination repair are involved in the error-free PRR pathway.
1.6.2. Assessing the Genes Comprising PRR
RAD6 and RAD18 were initially chosen as candidates for PRR studies, since mutations in both loci confer a UV sensitivity that is independent of a defect in NER (Cox & Game, 1974). Both rad6 and rad18 cells display an inability to resolve UV-induced single stranded DNA gaps in an NER- background
(Prakash, 1981), which is the defining characteristic of a defect in DNA damage tolerance. Other genes that belong to the RAD6 epistasis group were implicated in this pathway by epistasis analysis or by their involvement in mutation induction
(Friedberg et dl., 1995). The loss of UV-induced mutagenesis in rad6 and rad18 cells indicated that the mutagenic or error-prone DNA damage pathway exists and that it is controlled by the RAD6 pathway (Lawreiice & Christensen, 1976). The genes that provide the rnutagenic arm of the PRR pathway include REVI, REV3 and REV7. These genes were identified after isolating mutants defective in mutation reversion (Lemontt, 1971; Prakash, 1974). RADS (Johnson et al.,
1992), RAD30 (McDonald et al., 1997) and alleles of POL30 (Torres-Ramos et al., 1996) have been implicated in the error-free PRR pathway, and their placements were determined by epistasis analysis.
1.6.2.1. Rad6:RadlB Complex: DNA-damage Recognition?
Initially, RAD6 and RAD18 were seen to hction in the same pathway, indicating a genetic interaction. Coimmunoprecipitation studies provided evidence for the formation of a Rad6:RadM heterodimer (Bailly et al., 1994).
This same study indicated that Rad18 possesses a single stranded DNA binding ability, but Rad6 does not (Bailly et al., 1994). The complex possesses a ubiquitin-conjugating activity (discussed in section 1.6.2.2.), binds single stranded DNA and possesses an AïPase activity that is dependent upon binding (Bailly et al., 1997). The physical interaction between these proteins and their associated functions are required for PRR (Bailly et al., 1997b). The postulated purpose and function of heterodimer formation is similar to the function of RecA in E. coli ; the single stranded binding capabilities of Rad18 recognize single stranded DNA (formed by stalled replication forks), and target the Rad6 enzymes to the sites of damage (Bailly et al., 1997a). It has not been determined whether both Rad6 and Rad18 are required for the mutagenic response. Contradictory results for UV-induced mutagenesis have made the placement of Rad18 in either pathway difficult. Further careful analysis of the W-induced mutagenesis of rad18 in the SUP4-o system indicates that the mutagenic pathway of PN1 is RAD18 dependent (Cassier-Chauvat &
Fabre, 1991; Armstrong et al., 1994). It appears as though the rad18 and rad6 strains may enhance untargeted mutagenic events in post-irradiated cells
(Cassier-Chauvat & Fabre, 1991). How the Rad6:Radl8 complex functions within the PRR pathway has not been determined. The Rad6:RadlS complex appears to be only involved in PRR, since other cellular roles of Rad6 (discussed in Section 1.6.2.2.) are independent of RadlS, and rely on interactions with cellular proteins other than Radl8. It is not known what promotes the association between Rad6 and Radl8, but the dimer is suspected of DNA -damage dependent ubiquitination. Identification of the Rad6:RadlS target is crucial to better understanding the mechanisms of PRR.
1.6.2.2. Rad6: A Ub iqui tin-conjuga ting enzyme
The RAD6 gene encodes a protein involved in diverse cellular functions; rad6 null mutants are defective in error-free and error-prone PRR (therefore sensitive to a wide variety of DNA damaging agents) (Prakash et RI., 1993), sporulation (Morrison et nl., 1988), telomeric silencing (Huang et al., 1997), and protein degradation based on the amino-end rule (N-end rule) (Dohrnen et al.,
1991). Al1 of the cellular bctions catalyzed by Rad6 require its function as a ubiqui tin-conjuga ting enzyme; the use of si te-directed mutagenesis to substitute the necessary cysteine (Cys88) residue in the active site of Rad6 (Ubcî) confers the phenotypes of a rad6 null mutant (Sung et nl., 1990).
Ubiquitin-conjugating enzymes are an essential part of the ubiquitin pathway. The ubiquitin pathway functicns in protein modification; proteins are bound by the 76 amino acid ubiquitin (Ub) for functional modification, for cellular signalling and as a degradation signal (Fig. 2) (Reviewed in Dubiel &
Gordon, 1999). Ub-directed proteolysis is an essential process in the cell. It is required for the removal of abnormal proteins, for modification of protein function, and to modula te pro tein levels in tightly controlled processes (e.g. Protein Target
Protein Protein Degradation Modification
Figure 2: The ubiquitin pathway. 1. Ub is first activated by a ubiquitin- activating enzyme (El). 2. The activated ubiquitin is then bound to a target protein; target recognition and Ub binding is mediated by the ubiquitin- conjugating enzyme (EZ/Ubc) and the ubiquitin protein ligase (E3). Different forms of ubiquitin linkages have been identified, with the predominent form
(K48)marking proteins for degradation (3.a). The novel linkage (K63)is thought to be involved in protein modification (3.b). 4. Polyubiquitin diain formation is promoted by the E4 enzyme. cyclins in cell cycle progression) (Varshavsky, 1997). In protein degradation, proteins are targeted to the 26s proteosome after being bound by polyubiquitin chains (Varshavsky, 1997). The process of ubiquitination requires the activity of four different enzymes: a ubiquitin activating enzyme (El), a ubiquitin- conjugating enzyme (Ubc or E2), and in some circumstances a ubiquitin protein- ligase (E3) and a E4 enzyme (Hersko & Ciechanover, 1998; Koegl, 1999). The El enzyme activates Ub by binding the molecule to the El active site in an ATP- dependent marner; the E2 accepts the activated Ub ont0 a cysteine residue in its active site. The Ub-E2 complex then identifies the target protein based on
çpecific signal sequences, and the Ub is attached to the target protein
(Varshavsky, 1997). Identification of the target protein sometimes requires the cooperation of an E3 enzyme. The Ub molecules are added to the target protein as polyubiquitin chains, predominately linked through a lysine residue, Lys48
(Chau et al., 1989). Polyubiquitin chah formation requires the activity of the E4 enzyme which conjugates Ub moities to the growing Ub chah (Koegl, 1999).
The role of Rad6 is perplexing; how can a single activity be modulated to take part in such diverse pathways? The exhemely acidic carboxyl-terminal tail of Rad6 (20 out of 23 amino acids are acidic) mediates the ubiquitination of histone H2B (Sung et al., 1988; Robzyk et al., 2000), which is required for sporulation (Morrison et al., 1988). The amino-terminus of Rad6 is involved in protein degradation via the N-end rule (Dohmen et ai., 1991), in sporulation
(Morrison et al., 1988) and in error-free PRR (Watkinset al., 1993). Removal of the N-terminus of Rad6 (rad6,,,) does not disrupt Ub binding, but prevents the interaction behveen Rad6 and the E3 protein, Ubrl (Watkins et al., 1993). The Rad6-Ubrl interaction is required for N-end rule protein degradation. Interestingly, the rad6,,., mutation confers a moderate UV sensitivity (compared to rad6 nul1 mutant) and displays a UV-induced mutator phenotype (Watkins et al., 1993). It appears that the N-terminus of Rad6 may be involved in interactions that are specific for error-free PM. Ub contains seven lysine (Lys or K) residues. In a screen to determine whether other Ub linkages exist (aside from the essential Lys48-Lys76 Ub-Ub linkage), each of the Lys residues in Ub were mutated (Spence et al., 1995). The
Ub Lys633Arg63 (UbK63R)mutants were UV and MMS sensitive, but proficient in normal protein degradation (via the N-end rule) (Spence et al.,
1995). The UbK63R mutation is a partial suppressor to rad6 and confers a defect in UV-induced mutagenesis, but not spontaneous mutagenesis (Spence et al.,
1995). The mechanistic relationship between Rad6 and polyubiquitin chains linked via their K63 residue is still unknown. There is therefore extensive data on Rad6 activity, but more information is needed on how its E2 activity influences PRR, and what other enzymes are involved.
1.6.2.3. Translesional Replication
The two enzymes that define PRR (Rad6 and RadlB) do not contain activities that can account for the gap filling function required by the pathway.
Rad6 and Rad18 are necessary for PRR, but their function must be associated with activating or signalling other downstream PRIX events. Known enzymatic hctions that can be associated with gap filling are likely confined to recombinational and replication mechanisms. Three enzyme complexes have been associated with PRR activity; two represent DNA damage specific polymerases and one is postulated to represent error-free PRR. Of the polymerases, one is highly mutagenic (referred to as the mutagenic response) , while the other replicates over a damaged DNA template with surprising fidelity. The identification of these two hanslesional bypass activities does not
account for al1 the gap filling within PRR, but they are the only enzymes with an assigned gap filling function thus far.
1.6.2.3.1. The Mutagenic Response
Linking mutagenesis to the PRR pathway came with the observation that
both rnd6 and rad28 mutants display a defect in DNA damage-induced
mutagenesis (Lawrence & Christensen, 1976). RAD6 and RAD78 are not directly responsible for mutagenesis, but are implicated by their genetic interactions with REV3, REV7 and REVI. rev3, rea7 and revl mutants were isolated due to
their defect in DNA damage-induced mutagenesis (Lemontt, 1971; Prakash, 1974). The Rev3-Rev7 complex forms a non-essential DNA polymerase (Pol c). In vitro studies have demonstrated that Pol{ has the ability to bypass thymine
dimers; these lesions stall the essential polymerases such as Pola in the same
assay (Nelson et al., 1996a; Baynton et al., 1998). It is believed that Pol< provides
a cellular function analogous to the UmuC-UmuD' complex in E. coli, which is to bypass lethal DNA lesions not repaired by error-free repair processes, and to prevent ce11 death at the risk of mutation. Revl is a UmuC homologue that has an associated function with Pol 5; REVl encodes a deoxycytidyl transferase that can transfer dCTP at the 3' of a DNA primer (Nelson et al., 1996b). Revl is able to incorporate a dCTP across from an abasic site, which cm then be extended by
Po16 (Nelson et al., 1996b). The loss of mutagenesis in rev mutants, and the ability of Polc and Revl to replicate past lethal DNA lesions has defined REV3,
REV7 and REVl as the mutagenic pathway in S. cereuisine. Until recently, PRR was defined as the Rad6/RadlB mutagenic pathway.
Labeling the entire PRR pathway as mutagenic is misleading, since the data
indicates the error-prone pathway of PRR is not the preferred mechanism used by the ce11 (Paulovich et al., 1998). There are many experimental data that support a preferred error-free PRR mechanism. There is a contrast in the phenotypes between rad6 or rad28 mutants and rev3 mutants with regard to
DNA damage tolerance: rad6 and rad18 cells are extremely sensitive to DNA damaging agents such as W or MMS, but reu3 cells are only moderately sensitive (Lemontt, 1971). The difference in sensitivities could be accounted for by the existence of a Rad6-dependent error-free PRR pathway that is parallel to the Rev3-dependent mutagenic response. Furthermore, rad6 and rad18 cells exhibit an increased spontaneous mutation rate, where rev3 cells do not
(Cassier-Chauvat & Fabre, 1991; Lemontt, 1971).
1.6.2.3.2. Error-free Translesional Bypass: RAD30
Rad30 was isolated in a screen of the Saccharomyces Genome Database
(SGD) as a yeast protein that shares homology with the E. coli DinB protein
(McDonald et al., 1997). Rad30 also contains sequence homology with E. coli
UmuC and with S. cerevisiae Revl (McDonald et al., 1997). Disruption of RAD30 conferred a UV sensitive phenotype, causing the gene to be named as the other radiation sensitive mutants (rad). Initial studies indicated that rad6 and rad1 8 are epistatic to rad30, but the rev3 rad30 double mutants show a greater sensitivity compared to rev3 and rad30 single mutants (McDonald et al., 1997). The epistasis analysis combined with the observation that rad30 cells display a
W-induced mutator phenotype, indicated that Rad30 was part of the error-free
PRR pathway. In the past year exciting discoveries related to RAD30 resolved how
Rad30, UmuCD' and the Rad30 human homologue, hRad30, function in their respective organisms. As mentioned in section 1.4.2.2., the disease xeroderma pigmentosum is caused by a defect in NER, except for the XP-V complementation group. XP-V cells exhibit an inability to replicate past DNA damage (Lehmam et al., 1975). In 1999 two groups independently isolated the gene responsible for the XP-V phenotypes. The first group isolated the enzyme from normal celfs that could complement the XP-V defect. The enzyme identified is a novel, non-essential polymerase that cm replicate past DNA damage with amazing fidelity (Masutani et al., 1999a & 1999b). It was discovered simultaneously that the yeast Rad30 functions as a non-essential
DNA polymerase with the ability to replicate past cis-syn-thymine-thymine dimers, correctly inserting two adenines opposite the thymines (Johnson et al.,
1999). Further studies determined that the features of Rad30 that allow it to bypass pyrimidine dimers are responsible for the low fidelity and low processivity of Rad30 on undamaged DNA. Rad30 replicates undamaged DNA with a higher frequency of misincorporations, and camot replicate more than 6-
7 nucleotides before dissociating from the template (Washington et al., 1999).
The correlation between yeast Rad30 and XP-V was made by Masutani et al(1999b) when they observed sequence homology between their novel human polymerase (Polq) that could complement the XP-V defect, and yeast Rad3O.
Johnson, et al (1999) recognized the phenotypic similarities between XP-V cells and rad30 yeast cells, and screened a human cDNA Library using a sequence homologous to the N-terminus of the yeast RAD30, to clone hRAD3O. Once hRAD30 was identified and sequenced, the hhRAD30 locus in XP-V ce11 lines was analyzed. In each XP-V ce11 line, the hRAD30 gene was mutated, providing further proof that the missing activity in XP-V cells is fullfilled by
RadJO/Polr( (Johnson et al., 1999a).
Elwcidation of the biochemical function of Rad30 corroborates the earlier genetic interpretations. Rad30 can replicate past pyrimidine dimers, but its low processivity and low fidelity require the activity of another polymerase to continue DNA svnthesis without misincorporation (Washington et nl., 1999).
Both genetic and biochemical studies indicate that the Rad30-dependent translesional bypass is not the only error-free PRR pathway. Genetically, the
UV sensitivity of the rad30 reu3 double mutant does not match the extreme UV sensitivity of rad6 or rad28 (McDonald et al., 1997), suggesting that another error- free pathway exists fullfilhg the PRR role even when RAD30 and REV3 are defective. Biochemically, the Rad30 polymerase has only been shown to bypass pyrimidine dimers, but the RAD6 PRR pathway is responsible for the tolerance of a much more diverse set of lesions. The isolation and characterization of RAD30 has expanded our knowledge of DNA damage tolerance in yeast and hurnans, but only represents part of this important pathway. 1.6.2.4. Clarifying Error-Free PRR
1.6.2.4.1. Suppressors of rad6 and rad18 mutants
As previously mentioned, the rnd6 and rad18 mutants are extremely
sensitive to UV and to other DNA damaging agents (Cox & Parry, 1968). Suppressors (RADH/SRSZ) of the W sensitivity of both mutants were isolated
from separate screens (Lawrence, 1979; Aboussekhra et nl., 1989; Schiestl ef nl.,
1990). Functional suppression of a null mutant phenotype may provide
information on a genetic interaction between the suppressor allele and the null gene. The initiative to find suppressors for rad6 and rad18 was to uncover other genes involved in this pathway. The mutations (including radH) responsible for rnd6 and rad18 suppression are al1 allelic to the SRS2 gene. In another screen, srs2 was isolated independently from rad6 and rad18 s tudies as an allele responsible for an increased rate of gene conversion (Rong et nl., 1991). SRS2 encodes a DNA helicase with 3' to 5' polarity (Rong & Klein, 1993). Srs2 is believed to chamel lesions into the Rad6-dependent DNA damage tolerance pathway, preventing recornbination repair mechanisms from inappropriately resolving stalled replication forks.
Three lines of evidence support this hypothesis. First, the srs2 mutation will only suppress the extreme W or MMS sensitivities of rad6 or rndl8 if there is a functional recombination repair pathway (Schiestl et al., 1990). An additional mutation in RADSI, -52, -54, -55 or -57 results in a strain with a W sensitive phenotype similar to the single rad6 or rndl8 mutants (Schiestl et nl.,
1990). It is assumed that the extreme UV and MMS sensitivities caused by rad6 and rad18 mutations are partially due to the SrsZmediated inhibition of recombination repair mechanisms. The second piece of data that supports the
role of Srs2 as a molecular switch between the two pathways pertainç to the
phenotype of the srs2 single mutant. srs2 cells display a moderate W and MMS
sensitivity, for which suppressors have been isolated; the suppressors for srs2
are al1 allelic to RADS1 (Aboussekhra et al., 1992; Chanet et al., 1996). Removal of Rad51 in an srs2 background aborts potential use of homologous recombination, thus preventing unfavorable recombination events. The stalled
replication fork might not have structures that can be resolved using
recombination mechanisms. A third example supporting the view that Srs2 behaves as a enzyrnatic switch is the observation that srs2 cells have elevated
levels of recombination. It is argued that in the absence of SRS2 function, cells promote recombination based ENA damage tolerance (Rong et al., 1991).
Studies pertaining to SRSZ, RAD6, RAD18 and the genes of the
recombination repair pathway have provided genetic evidence for the organization of PM. Evidence for Srs2 promoting Rad6-mediated PRR is phenotypic, not biochemical. There are two possible hypotheses for the three
above mentioned observations. In the stalled replication fork, Srs2 may catalyze
the unwinding of nascent DNA, creating a template for Rad6-mediated PRR.
Altematively, Srs2 may bind to a DNA structure within the replication bubble and, via protein:protein interactions, preferentially initiate the Rad6 pathway.
Until more biochemical data is obtained, the process by which Srs2 promotes Rad6 PRR is unknown. 1.6.2.4.2. Other Candidate Genes in Error-Free PRR In a screen to identify "reversionless" mutants, alleles of rm3, and an allele of revl and reu2 were isolated (Lemontt, 1971). RE Vl and REV3 are essential for the mutagenic pathway of PRR. reu2-1 is allelic to RAD5 and has been placed within the RAD6 epistasis group (Johnson et al., 1992). Within the RAD6 pathway, MD5has been assigned to the error-free sub-pathway, as the rad5 reu3 double mutant has a level of UV-induced killing that is synergistic with the UV- sensitivity of the corresponding single mutants (Johnson et al., 1992). The initial isolation of rev2-l for its "reversionless" phenotype was possibly an artifact of the allele, since the reduced suppression (of the ochre mu ta tion) was specific for this ochre arg4-17 mu ta tion (Lemontt, 1971). More recent investigations have observed that a rad5 nul1 mutation has no effect on
UV-induced mutagenesis, therefore influencing error-free rather than error- prone repair (Johnson et al., 1992). Mutations in rad5 confer an increase in simple repeat stability, and an elevation of non-homologous end-joining for the repair of DSBs (Johnson et al., 1992; Ahne et nl., 1997). Within Pm, rnd18 and rad5 mutants are hyper-recombinant, and the level of recombination of the rad5 rad18 double mutant does not exceed levels seen in either single mutant
(Liefshitz et al., 1998). The hyper-recombinant phenotype in rad5 and rad18 cells is dependent upon the RAD52 pathway (Liefshitz et al., 1998). Spontaneous mutagenesis in both rad5 and rad18 mutants is increased, but the levels in the double mutant are depressed (Liefshitz et ai., 1998). Biochemical analysis indicates that RAD5 encodes a 134 kDa single-stranded DNA-dependent
ATPase, containing a RING-finger motif (for DNA binding), a leucine zipper motif, and domains found in helicases (Johnson et al., 1992; Johnson et al., 1994). Despite the presence of consewed motifs found in helicases, Rad5 does not
contain helicase activity (Johnson et al., 1994). Further biochemical and genetic
data is required to functionally place Rad5 within Pm. The activities of Rad5 appear to function solely for PRR. During the past twenty years, there has been a gap in the biochemistry of error-free PRR due to
unsuccessful attempts to identify other genes involved in error-free PM.
Recently, the isolation of two mutant alleles has added new candidates for error- free PRR. Both POL30 (yeast PCNA homologue) and POL2 (polymerase 6) are
essential for replication, and by the isolation of mutant alleles, a comection between the replication rnachinery and PRR was constructed.
Using alanine scanning mutagenesis, Ayyagari, et cil (1995), isolated po130 mutants with various phenotypes. The po130-46 allele was created by combining
two of the above double mutants, creating a quadruple mutant (four charged
amino acids were substituted by alanines). poZ30-46 is especially interesting
shce it is sensitive to UV and MMS and does not appear to have a defect in replication (Ayyagari et al., 1995). To assess whether the diverse drug sensitivity of po130-46 was due to a defect in PRR, epistasis analysis was performed. rad18 and rad6 are epistatic to po130-46, while rm3 and po130-46 display a synergistic killing response (Torres-Ramoset al., 1996). As seen in rad5 and srs2 mutants, po130-46 has wild type levels of UV-induced mutagenesis (Torres-Ramos et al.,
1996). Torres-Ramos, et al (1996) were able to confirrn a role of the PCNA homologue in PRR. In an alkaline sucrose gradient assay, W-irradiated po130- 46 mutants have increased fragmentation of genomic DNA, even after a recovery period (Torres-Ramos et al., 1996). The isolation and characterization of pol30-46 was very important, since it was the first error-free PRR mutant to display a PN1 defect, and the PRR defect detected in po130-46 but not rev3 indicates the preference of error-free over error-prone PRR. The involvement of Po130 in PRR confirmed the existence of the error-free pathway, but did not help resolve the mechanism of translesional bypass. Po130 is involved in many cellular activities, including DNA replication, cell-cycle control, NER and mismatch repair (Jonsson & Hubscher, 1997). Both of the polyrnerases Po16 and Pole require the activity of Po130 aonsson & Hubscher,
1997). By irnplicating Po130 in error-free PRR, both Pol6 and Pole were candidates for providing the error-free translesional bypass mechanism.
Temperature sensitive alleles of POL3 (Po16) and POL2 (Pole) were used to assess the involvement of both polymerases. POL2 is an essential gene, but its cellular role has not been elucidated. The pol2-18 mutant did not display a defect in PRR in the alkaline sucrose gradient assay, indicating that it is not involved in error-free translesional bypass (Torres-Ramos et al., 1997). POL3 is also an essential gene, encoding the catalytic subunit of Po16 polymerase, the
DNA polymerase required for leading and lagging strand elongation (Newlon,
1996). The poi3-3 allele of POL3 has a PRR defect, which is detected on an alkaline sucrose gradient after UV-irradiation (Torres-Rarnos et ai., 1997). Unlike po130-46, little data has been gathered on po13-3 pertaining to PRR. Until more data on the po13-3 allele is available, no further analysis can be performed.
Giot, et al (1997) isolated another POL3 temperature sensitive allele with PRR defective phenotypes. po13-13 is sensitive to a wide variety of DNA damaging agents, and is synthetically lethal if combined with a rad50 or rad52 mutation, indicating that recombination could replace the function missing in po13-13 cells (Giot et al., 1997). Epistasis analysis places po13-13 within the RAD6 pathway (Giot et al., 1997).
Suppressors of the po13-13 allele were isolated (sdp5-2 and sdp5-15), and are allelic to the HYS2 /POL31 gene (Giot et al., 1997). POL31 encodes an essential subunit of Pol6 that forms a heterotetramer with another subunit, Po132
(Gerik et al., 1998). POL31 was initially isolated in a screen for mutants sensitive to hydroxyurea, indicating a defect in nucleotide metabolism, replication, cell- cycle checkpoint control, or in DNA damage repair (Gerik et al., 1998). The phenotypes associated with the above-mentioned po131 alleles link this gene with PRR. The other subunit of the Po16 holoenzyme, Po132, may also be linked to PRR. POL32 encodes a non-essential subunit of Pol8 that, when mutated, confers a DNA damage and hydroxyurea sensitive phenotype (Sugimoto, 1995;
Gerik et al., 1998). The lack of epistasis analysis prevents the assignment of
POL32 to any DNA repair pathway, but the reduction in spontaneous and DNA damage-induced mutagenesis observed in the mutants implicates Po132 in a mutagenic response. Po132 also interacts with Po130 (Gerik et al., 1998). The sensitivity of po132 to a wide variety of DNA damaging agents, and the involvement with Po130 (involved in error-free Pm) makes Po132 another candidate PRR gene.
It is well established that S. cerevisiae contaiw a DNA tolerance system, but it has not been clearly determined how the assigned PRR enzymes cooperate to resolve single sbanded gaps formed by interrupted replication forks. Early models of PRR separate recombination processes from Rad6-dependent PRR, advocating that recombination-based mechanisms of DNA damage tolerance are prevented by Srs2 (Fig. 3) (Friedberg et al., 1995). To better understand how Interruption of S-phase Replication Fork
RAD6-Dependent PRR Recombination Repair
OPo130
Figure 3: Postreplication repair overview. Ln the event of a stalled replication fork, the RAD6-dependent DNA damage tolerance pathway is the preferred pathway to resolve the resultant single stranded gaps. The presence of the Srs2 helicase prevents the recombination repair pathway from processing any aberant replication structures. Many enzymes have been genetically linked to
RAD6-dependent PRR, but their mechanisms are unknown. Rad6-dependent PRR prevents the lethal consequences of DNA damage, resezrch must focus on two areas of PRR. First, more biochemical information must be gathered on the enzymes proposed to function within PM. Second, more enzymatic functions must be associated with the Rad6-pathway either through identifying new enzymatic potentials in known enzymes or by the isolation of novel PRR genes. This second research strategy precedes the rationaie of this dissertation.
1.7. MMSZ: Rational for this Project
To prevent the pathological consequences of DNA damage, researchers attempt to better understand how DNA repair pathways recognize and resolve the damage. Identification of novel DNA repair enzymes most often occurs through the isolation of mutant cells that are sensitive to a specific fom of DNA damage (ie. 'JV).Prakash & Prakash (1977) irradiated a yeast culture, then subsequently screened to specifically isolate mutants sensitive to alkylation damage (MD strains). Five MMS-sensitive mutants were isolated in the screen and the mutant alleles were designated as rnrnsl-1, nzms2-1, ntms4-1,1nrns5-2 and rnnzs22-1 (Prakash & Prakash, 1977).
The aim of my project was to identify and characterize the gene whose mutation was responsible for the MMS sensitivity in the yeast strain MD-2
(containing the nzrns2-1 allele). Using the MD-2 strain isolated by Prakash and
Prakash (1977),1 was able to clone MMS2. Through the characterization of
MMS2, another member of the PRR pathway was discovered. In determining the genetic relationships between MMSZ and other genes associated with PM, the notion of an error-free PRR subpathway was confirmed. Human homologues of MMS2 have been isolated and are currently under investigation. Understanding PRR in yeast may provide a mode1 to better understand DNA damage tolerance processes in humans. Collaborative research efforts on yeast MMS2 and its human homologues may produce information that might be of use or interest to those in the field of disease prevention. CWAPTER TWO: MATERIALS AND METHODS
2.1- Yeast Genetics
2.1.1. Yeast Strains and Cell Culture
The yeast strains used in this study are listed in Table 1. WX17-4a was isolated from MD-2/FY86 diploid segregants in order to combine mms2-l with zlrn3 for library screening. The isogenic rnd4d, rnd6A, rndlBA, rnd50A and rev3A mutants were described previously (Xiao et al., 1996). The URA3 selectable marker of WXY9221 was removed by selection on a 5-fiuoroorotic acid plate
(Boeke et ni., 1987) to obtain WXY9579. The haploid strain T43 bearing an ntrns2A mutation and its corresponding wild type, BY448, were gifts from Dr. S. Bacchetti (McMaster University, Canada). Yeast cells were grown at 30°C either in a rich YPD medium or in a synthetic glucose (SD) medium supplernented with amino acids and bases at recommended concentrations (Sherman et ni.,
1983).
ïhe PY39-0 and PY39-46 strains were obtained from Dr. P. Burgers
(Washington University School of Medicine, St. Louis MI). See Table 1 for the genotypes and modifications to the PY39 background. Strain XS-803-3A was a gift from Dr. D. Gietz (University of Manitoba, Winnipeg). Table 1. Saccharomyces cerevisine s trains
Strain Genotype Source
8-635 n cycl-225 lys2 hisl trp2 F. Sherman
MD-2 B635 with nims2-l L. Prakash
FY86 a Iiis3-A200 il rd-52 leu 2-Al GAP F. Winston
WX17-4a n liis3-A200 ir ra3-52 lys2 ?iztns2-1 This stud y
DBY747 a his3-41 1eu2-3,112 QI-289 iirn3-52 D. Botstein
SBU DBY747 with nlms2::URA3 This study
SBL DBY747 wi th rrirns2::LEUZ This study
WXY9579 DBY747 wi th rnd50A::hisG Xiao et al (1996)
SBU5Oh DBY747 wi th rad50A::hisG mnls2::URA3 This shidy
WXY9394 DBY747 with rnd4A::hisG::URA3::hisG Xiao et al (1996)
PY39Ar5As2 PY39-0 with rad5::hisG-URA3-hisG srs2A::LEUZ This study
PY39-46Am2 PY39-46 with nlrns2A::LEU2 This study
PY 39-46b2 PY39-46 with srs2A::LEU2 This study b
PY 39-46Ar5 PY39-46 with rad5:: kisG-URA3-hisG This study
PY39-46Ar5As2 PY39-46withrd5::hisG-URA3-lrisCsrs2A::LELI2 This study
XS-803-3A MAT a leu2-3,112 ura3-52 his-l canls trp2 Dr. D. Gietz
XS-803-3A m2A XS-8033A with nzms2::LEU2 This s~dy SBL-T DBY747 mms2::LEU2 telomere::URA3 This study
______CC______------"-"-~"----~------"------"------The pADH4UCA-N vector created by Dr. D. Gottschiing (Fred
HutdUnson Cancer Research Center, Seattle WA) was digested with San and
EcoRI to release the adh4::URAJ cassette. The yeast strain SBL was hansformed with the above cassette. Ura+ colonies were selected and Southem analysis was used to confirm that the URA3 gene had properly integrated into the telomere of chromosome VII.
YPD is a standard, complex medium composed of 1% Bacto-yeast extract,
2% Bacto-peptone, 270 glucose, and 2% agar (if making YPD plates). YPGal media replaces the 2% glucose with 2% galactose. Synthetic Dextrose minimal medium (SD medium) is used for selective growth of yeast auxohophs. It contains 0.67% yeast nitrogen base (without amino acids), 2% glucose, 2% agar
(if making SD plates), and addition of any necessary auxotrophic suppliments.
If a necessary nutritional supplement (X) is excluded the media is denoted as
SD-X. Synthetic Complete (SC) medium contains 0.67% yeast nitrogen base
(without amino acids), 2% glucose, and 2% agar (if making SC plates), but requires the addition of 0.2% drop-out mix. Drop-out mix provides the media with growth requirements commonly needed for the standard auxotrophies in lab strains of yeast. Drop-out mix: 0.5 g adenine, 2.0 g alanine, 2.0 g arginine, 2.0 g asparagine, 2.0 g aspartic acid, 2.0 g cysteine, 2.0 g glutamine, 2.0 g glutamic acid, 2.0 g glycine, 2.0 g histidine, 2.0 g inositol, 2.0 g isoleucine, 2.0 g lysine, 2.0 g methionine, 2.0 g phenylalanine, 2.0 g proline, 2.0 g serine, 2.0 g threonine, 2.0 g hyptophan, 2.0 g tyrosine, 2.0 g uracil, 2.0 g valine, 10.0 g leucuie, and 0.2 g para-aminobenzoic acid. Any of the above nutrients in the drop-out mix can be removed to provide a selection media for yeast transformation. YM-1 medium is composed of 5.0 g yeast extract, 10.0 g peptone, 6.7 g yeast nitrogen base (without amino acids), 0.01 g adenine, 0.01 g uracil, 10.0 g succinic acid, 6.0 g sodium hydroxide, and 10.0 g glucose in a total volume of 1 L, with a pH of 5.8. For long term storage, yeast cells are grown on plates (rich or minimal media) at 30°C. After 2-3 days growth the yeast cells are removed from the plate with a sterile tooth-pick, and inoculated into 1.0 ml of sterile 15% (v/v) glycerol.
The cells are then stored at -70°C.
2.1.2. Yeast Transformation
Yeast cells were transformed using the method described by Hill et nl (1991). 5 ml culture(ç) of S. cereuisiae were grown ovemight at 30°C in rich media (unless otherwise mentioned). The next day 200 pl of culture was sub- cultured into 30 ml of fresh media, and allowed to grow at 30°C until it reached a mid-logarithrnic phase of growth. The yeast cells were collected by centrifugation, washed in a LiOAc solution (0.1 M LiOAc, lOmM Tris HCl (pH
&O), ImM EDTA) and resuspended in 1 ml of LiOAc solution.
For each transformation, 100 pl of cells was mixed with 4 pl of carrier DNA
(single stranded salmon sperm) and 1-5 pl of transforming DNA in a 1.5 ml centrifuge tube. After a 5 minute incubation at room temperature, 280 pl of
PEG,, solution [50% polyethylene glycol(MW=4000) in LiOAc ] was added and the contents mixed by inverting 4-6 times. After the transformation mixture was incubated for 45 minutes at 30°C, 39 pl of DMSO was added, followed by a 5 minute heat shock in a 42'C waterbath. Following the heat shock treatrnent, the transformation mixture was washed with double distilled water (ddH,O), then resuspended in 1 ml of ddH,O. The resuspended cells are plated onto the appropriate minimal media. 2.1.3. Yeast Plasmid Extraction
The protocol for the extraction of yeast DNA was developed by Hoffman and Winston (1987). Yeast cells are obtained from plates or from a liquid culture. Cells grown on plates were scraped with a sterile toothpick and resuspended in 1.5 ml eppendorf tube containing 200 fl of extraction buffer [2%
TritonX-100,1% SDS, 0.1 M NaC1,l mM EDTA, 0.01 M Tris HCI (pH8.0)]. Cells from liquid culture were collected by centrifugation and resuspended in 200 pl of extraction buffer. 0.1 ml of phenol, 0.1 ml of chloroform and 0.3 g of acid- washed glass beads were added to the ce11 mixture, and the tube is vortexed for
2 minutes (at top speed). After a 5 minute centrifugation, the top aqueous layer was hansferred to a clean tube, and precipitated by 2 volumes of 100% ethanol.
When a pure DNA preparation was required, the precipitated DNA was resuspended in 200 pl of TE (10mM Tris HCI, 1mM EDTA, pH $.O), then treated - with 5 @ of RNase (10mg/ml stock) at 37°C for 10 minutes. After the Nase heatment, the DNA was precipitated in 2 volumes of 100% ethanol with 8 pl of
5M NaCl.
2.1.4. Yeast Genomic DNA Isolation
To isolate genomic DNA from S. cereuisine for Southem blots a modification of the protocol described in Hoffman & Winston (1987) was used.
For the isolation of genomic DNA the ce11 mixture was vortexed for three minutes rather than 2 minutes. Aside from the difference in vortex tirne, the entire isolation is identical to the protocol in Section 2.1.3. 2.1.5. Screening a Genomic Library
A YCp50-based yeast genomic library (Rose et al., 1987) was obtained from Dr. M. Rose (Princeton University) and used to transform WX17-4a. A two-step screening protocol was followed. First, Ura' transformants were obtained on SD-Ura plates. A total of approximately 10,000 Ura' colonies were individually streaked ont0 YPD and YPD+0.04% MMS. The MMS-resistant clones were subjected to a plasmid CO-segregationtest and the YCp50-based plasmids were recovered by transforming Escherichia coli with the extracted yeast total DNA.
2.1.6. Co-Segregation Test
This test was used to assess whether the isolated clone was responsible for restoring the MMS resistant phenotype in WX17-4a, by determining the phenotype of cells with and without the plasmid. The YCp50 plasmids used to cowtruct the library contain the URA3 marker gene.
Yeast cells were inoculated into 2 ml of YPD liquid media, and grown overnight at 30°C. 10 pl of the saturated culture was subcultured in 2 ml of fresh
YPD liquid media, and incubated overnight at 30°C. These cells were diluted and plated to obtain plating dilutions of 104 and 10". After a 2 day incubation at
3O0C,colonies were selected and replica-plated ont0 minimal media (SD media) including (+Ura) or excluding (-Ura) uracil. Both Ura' and Ura' cells were selected, and streaked ont0 selective MMS plates. Ura+,MMS resistant colonies are an indication of plasmid-dependent drug resistance. 2.1.7. Drugs and Special Media
Methyl methane sulfonate (MMS) was purchased from Aldrich
(Milwaukee, USA), as an aqueous solution. To prepare MMS dmg plates, MMS was added irnmediately before plating to reduce the amount of drug degradation. CNitroquinoline N-oxide (4-NQO) was purchased from Sigma (St.
Louis, MO). The stock solution is made in acetone at a concentration of 10 mg/ml. Fluoro-orotic acid (FOA) is toxic to yeast cells with a functional URA3
(Boeke et al., 1984). The gene product of URA3 metabolizes the FOA to a toxic intermediate. FOA was purchased frorn US Biologicals (Swampscott, MA), and is stored as an aqueous solution at -20°C. FOA plates were prepared as follows:
A mixture of 0.67% Bacto-yeast nitrogen base, 0.2% Drop-out mix (-URA), 2% glucose, 50 pg/ml uracil, 0.l0/0 5-FOA in 500 ml of distilled water was filter sterilized. The above solution was combined with 500 ml of an autoclaved 2%
Bacto-agar, then poured into petri dishes.
2.1.8. Ce11 Killing
MMSinduced liquid killing was performed as previously described (Xiao et al., 1996). Briefly, cells were grown ovemight in YPD media; 500 pl of ovemight culture was subcultured into 5 ml of fresh WD, and allowed to grow until a mid-logarithmic phase of growth was achieved. 1 ml of culture was removed, diluted and plated on YPD with a plating dilution of 10'~;these cells represent the untreated control cells. At this time, MMS was added to the rernaining culture, to final concentrations as indicated in each experiment. At given time intervals (up to 1 hour), 1 ml of cells were removed, washed twice with sterile water, diluted to an appropriate concentration and plated in duplicate on YPD plates. Plates were incubated for 3 days at 30°C. For W treatment, cells were plated at different dilutions and then exposed to 254 nm UV light, either in a UV crosslinker (Fisher Sci. model FB- WXL-1000 at ca. 2,400 pW/cm2) or with a UV lamp (UVP model UVGL-25 at 40 pW/cm2) at specified doses. Cells were plated in duplicate on YPD to score ce11 survival, and the plates were incubated at 30°C for 3 days. Irradiation and the subseqent incubation of the cells was performed in the dark, to prevent photoreactivation.
2.1.9. Spontaneous Mutagenesis Assay
DBY747 bears a trpl-289 amber mutation that can revert to Trp' by several different mutational events. Spontaneous Trp' reversion rates of
DBY747 derivatives were measured by a modified Luria and Delbruck fluctuation test as described (Von Borstel, 1978). An ovemight yeast culture was used to inoculate five tubes, each containing 10 ml of fresh WD, to a final titer of 20 cells/ml. Incubation was continued until a titer of 2x108 cells/ml was reached. Cells were collected, washed, resuspended and plated. Each set of experiments contained five independent cultures of each strain; each culture was plated ont0 YPD in duplicate to score total survivors, and ont0 SD-Trp plates to score Trp' revertants. Spontaneous mutation rates (number of revertants per ce11 per generation) were calculated as previously described
(Williamson et al., 1985). To calculate the frequency of spontaneous mutagenesis the following formula was used:
Frequency (F)= total ce11 number of TRP' cells total number of viable cells To calculate the rate of spontaneous mutagenesis, the following formula was used:
Rate = 0.4343 x Freauency log(tota1 ce11 number) - log(initia1ce11 number)
The formula was derived to determine mutation rate for a replicating system, where 0.4343 logarithm of e.
2.1.10. UV-Induced Mutagenesis Assay
As mentioned above, DBY747 bears a trpI-289 amber mutation that can be reverted to Trp' by several different mutational events. An adaptation of the method described by Watkins et al (1993) was used to assess the UV-dependent reversion of this TRPl allele. The cells were grown overnight at 30°C in 3 ml of
YPD media. 100 pl of the ovemight culture was sub-cultured into 5 ml of fresh
YPD media. When the culture reached mid-logarithmic growth, serial dilutions were made to obtain ce11 numbers that could be quantitated on YPD plates. Undiluted cultures were plated onto SD-trp plates to detect for TRP' reversion.
Using a W lamp (UVP mode1 UVGL-25 at 40 pW/cm2), both the YPD and SD- trp plates were exposed to 2,4, and 6 J/m2, in the dark. The plates were then incubated at 30°C in a dark chamber for three days to prevent photoreactivation.
2-1-11.Assessing Telomeric Silencing
Telomeric silencing cm be monitored by the qualitative observation of gene expression, when the marker gene is placed close to a telomere. Dr. D.
Gottschling (Fred Hutchinson Cancer Research Center, Seattle WA) created a integration cassette, which specifically integrates the URA3 gene into the telomere of chromosome W (Gottschling et al., 1990). The vector containing the
integration cassette (Table 3) was digested with Ça11 and EcoN restriction endonucleases (New England Biolabs, Bethesda MA, USA) to release the fragment. The SBL yeast strain was transformed with the SnlI-EcoRI fragment, and transformants were selected on SD-Ura plates. Southern analysis was used
to confirm URA3 integration into the telomere of chromosome W, creating strain SBLtel:: URA3 (Table 1).
Serial dilutions were made of an overnight SBLtel::URA3 culture, a
TLCLhuman tel::URA3 culture (negative control), and a DBY747 culture
(positive control). The cells were replica plated ont0 YPD and FOA plates. The inability of cells to grow on FOA plates is taken as an indication that they actively banscribe URA3, and therefore do not exhibit telomeric silencing.
2.1.12. Protein Degradation Assay
The pUB23-x (pUb-M, -L, or -R) vectors were transformed in rad6, mms2 and
rrbcl3 mutants of both FY86 and 10A backgrounds (Section 2.1.2.). Ura+colonies were selected and re-streaked ont0 SD-Ura plates, to ensure a Ura' phenotype. Al1 the transformants were grown overnight in SD-ura liquid medium (3.0 ml); the next day, the cells were subcultured in 3 ml of SGal-Ura and again incubated overnight at 30°C. 1 ml of ovemight culture was removed to determine the ce11 concentration using OD,. 100 pl of culture was combined with 100 pl of Buffer Z (60 mM
Na,HPO,, 38 mM Na,H,PO,, 10 mM KCl, 1 mM MgS04-7Hz0). To permeablize cells, 50 ul of 0.1% SDS and 3 drops (50 pl) of chloroform were added and the tubes were vortexed at top speed for 10 seconds. To induce cells, 200 pl of 4 mg/ml
ONPG was added, and the mixture was incubated at 28OC until the yellow colour appeared (30 minutes). The reaction was terminated by adding 500 pl 1M Na2C03; the ce11 debris was removed by centrifugation at 5K for 5 minutes. 1 ml aliquots of supematant were removed to determine P-gal activity at 0D420(use Buffer
Z+SDS+chloroform for blank). Specific activity was calculated as follows:
Specific activity= 1000 x (OD42n,,)- where t= time and v=volume. (OD600nrn)(t) (VI
2.1.13. Sponilation Assay
Haploid strains FY86 mms2::LEUZ (MATa)and XS-803-3A mms2::LEUZ
(MATa), were grown on rich media (YPD) plates to obtain actively growing cells. The above strains were chosen based on their auxotrophic differences
(FM6 strain has a his3-A200 mutation, and XS-803-3A has a his2-l mutation; the resultant diploid produces a His' phenotype), providing a method to detect diploid formation. The strains were cross-streaked in an X-formation on SD-His plates, mixing the cells a t the centre of the X (where the two strains intersect).
After rnixing, the centre cells were streaked out on the surface of the plate to obtain single diploid colonies. The plates were incubated for 2-3 days at 30°C to select for individual colonies.
Once His' diploid colonies were obtained the cells were inoculated into 3 ml of pre-sporulation media (0.8% Bacto yeast extract, 0.3% Bacto peptone, 1O0/0 dextrose), and the culture was incubated ovemight at 30°C. 10 pl of the ovemight culture was spotted ont0 sporulation plates [1% potassium acetate,
0.1% Bacto yeast extract, 0.05% glucose, ( 2% agar)], and was incubated for 2-3 days, until the spot had grown. A hemacytorneter was used to monitor and quantitate the formation of tetrads under the microscope. 2.2. Molecular Biology Techniques
2.2.1. Bacterial Culture and Storage
The E. coZi strains commonly used for bacterial transformation are DH5a and DHlOB (GibcoBRL, Grand Island, N'Y USA). Transformed strains were stored on LB plates (1% Bacto-tryptone, 0.5% Bacto-yeast exhact, 0.5% NaCl and
1.2% Agar) containing 50pg/ml of Ampicillin, shce al1 of the vectors used contained the bln marker gene. For long term storage of transformed cells, cells were grown ovemight in 900 pl of LB + Amp; the following day 100 pl of DMSO was added and the cells were immediately placed in a -70°C freezer.
2.2.2. Plasmid DNA Isolation (Mini-prep)
Plasmid amplification and isolation was performed as described in Maniatis et al. (1982). Cells were inoculated into LE3 + Amp liquid media, and grown ovemight at 37°C. Cells were collected by centrifugation, the supernatant removed, and the pellet was resuspended into 350 pl of bacterial plasmid prep solution (8% sucrose, 0.5% TritonX-100,50 mM EDTA pH 8.0,10 rnM Tris-HC1 pH 8.0). After adding 25 fl of lysozyme (10 mg/ml; Sigma, St.
Louis MI), the mixture was put in a boiling water-bath for 1 minute, then centrïfuged for 10 minutes. The resultant pellet was removed, and the 8 pl of
5M NaCl and 2 volumes of 100% ethanol were added to precipitate the DNA. 2.2.3. Large Scale DNA Isolation (Maxi-prep)
A 5 ml culture of the transformed bactenal cells was incubated over night at 37'C. The culture was then subcultured into 500 ml of fresh LB + Amp medium and incubated until an ODmm reached 0.6. At this stage, 2.5 ml of chloramphenicol(0.033 g/ml in alcohol) was added and the culture was further incubated ovemight at 37'C. Cells were collected by a 10 minute centrifugation at 6700 rpm in a Beckrnan GSA rotor, and resuspended in 12 ml of sucrose solution (10% sucrose w/v, 50 mM Tris-HC1, pH 8.0). The cells were transferred to a 40 ml centrifuge tube, and the previous containers were rinsed with an additional 5 ml of sterile water to collect any residual culture. 2 ml of lysozyme
(10 mg/ml) and 2.4 ml of 0.5 M EDTA (pH 8.0) were added to the cells, the contents mixed, and the tube was incubated on ice for 10 minutes. 1 ml of 2% sarkosyl was added and the tube was mixed to lyse the cells. The viscous ce11 solution was transferred to a 30 ml Corex centrifuge tube and centrifuged for 70 minutes at 12000 rpm in a Beckman 5534 rotor. The supematant was collected in a 40 ml plastic centrifuge tube, and 8 ml of both phenol and chloroform were added. The tube was centrifuged for 5 minutes at 5000 rpm (Beckrnan 5534 rotor) and the supematant was transferred to a clean tube, where another 16 ml of chloroform was added. After another 5 minute centrifugation the supematant was removed and divided evenly into 2 glass centrifuge tubes; 2 volumes of 100% ethanol were added to each tube, they were covered with parafilm, then placed in the -20°C freezer for at least 3 hours.
The tubes were centrifuged at 10000 rpm (Beckrnan SS34 rotor) for 20 minutes, and the tubes were inverted to dry the pellets. Each pellet was dissolved in 9.5 ml of TE (10 mM Tris-HC1,l mM EDTA, pH 8.0 ), combined, then mixed with 9 g of cesium chloride. The mixture was added to heat sealer
tubes using a 10 ml syringe, 0.35 ml of ethidium bromide (10 mg/ml) was
added, and the volume was adjusted with a solution of TE with cesium chloride.
The tubes were heat sealed, and centrifuged in the Beckrnan Ultracentrifuge for 16 hours at 55000 rpm.
Following the centrifugation, the tubes were removed and the DNA band
of interest was detected using a UV lamp, and was exhacted using a syringe. The ethidium bromide solution was exhacted mice with equal volumes of butanol, and the aqueous layer was dialyzed in a large volume of TE (with 3
changes of TE). The DNA was ethanol precipitated, resuspended in 500 pl of
TE, and quantitated at OD,,,.
2.2.4. Preparation of Competent Cells
For chernical transformation, E. coli DH5a or DHlOB were treated as
described by Chung et al., (1989). Cells were grown in LB to an OD,,, of 0.3-
0.4. The cells were subsequently diluted 1:l in TSS solution (lx TSS: LB with 10% PEG,, 5% DMSO, and 50 mM Mg+' (MgSO, or MgCl), pH 6.5). Cells were
aliquoted, 50 pl/tube, and quickly placed in -70°Cfor storage.
For transformation by electroporation, E. coli cells were prepared as
suggested in the BioRad E. coli Pulser manual. 1 litre of culture was incubated
until an OD, of 0.6 was reached. The culture was centrifuged at 3500 rpm in a
Beckman GSA rotor and the pellet was resuspended in 500 ml of 10% sterile
glycerol. The centrifugation was repeated 4 times, with each pellet resuspended
in a reduced volume; the last pellet was resuspended in 4 ml of cold, sterile 10 '10 glycerol. The cells were aliquoted into 1.5 ml centrifuge tubes to a volume of 25 pl, and were quickly placed in the -70°Cfreezer for storage.
2.2.5. Bacterial Transformation
2.2.5.1. Chemical Method
Each transformation used a 1.5 ml eppendorf tube containing 50 @ aliquots of E. coli cells, which were competent for chernical transformation (See section 2.2.4.). The volume of transforming DNA added to the cells was no greater than 10% of the final volume. After 30 minutes on ice, the cell-DNA mixture was heat shocked for 1 minute at 42'C. Following the heat shock, 450 pl of SOC media ( 2% Bacto-tryptone, 0.5% Yeast extract, 10 mM NaCI, 20mM
MgCl,, 20mM MgSO,, 20mM glucose) was added and the cells were incubated at 37'C for 1 hour. Usually, 100 pl of the transformation solution was plated ont0 LB+Amp (50pg/ml of Ampicillin, final concentration) plates, depending on the expected efficiency of transformation.
2.2.5.2. Electroporation Method
The DNA to be transformed was added to E. coli cells competent for electroporation. After a brief incubation on ice, the ce11 mixture was transferred to a chilled electroporation cuvette (BioRad), where the cells were exposed to a voltage of 2.5 kV or 1.8 kV (for cuvettes with 0.2 or 0.1 mm width, respectively) using the E. coli Pulser (BioRad). After electroporation 280 pl of SOC was added to the cuvette, and the ce11 mixture was transferred to a 1.5 mi centrifuge tube for a 45 minute incubation at 37°C. Following the incubation, the cells were plated on LB + Amp plates and incubated at 37'C ovemight.
2.2.6. DNA Sequencing
Nucleotide sequences of the MMS2 open reading frame and its surrounding regions were determined by a dideoxy chah terminating method
(Sanger et al., 1977) using a T7 DNA Polymerase Sequencing kit (Pharmacia
LKB; Little Chalfont, Bucks UK). The gel was electrophoresed in 1 x TBE buffer
(890 mM Tris, 890 mM boric acid, and 20 mM EDTA, pH 8.0).
A combination of short and long electrophoresis times were used to sequence the insert in the YCp-MMS2 clone. The Ml3 universal primer and the reverse primer were used to determine the sequence of both strands of the insert
(listed in Table 2). For accurate analysis of the interna1 sequence, deletions were created in the pTZ-MMS2 clone (BstBI deletion, EcoR1-EcoRV deletion, and a
BglII deletion, to create pTZ-MMS2a. Sequencing of the deletion construct used both Ml3 universal primer and the reverse primer. After obtaining the MMS2 sequence (GenBank accession number U66724), it was analysed to search for intron sequences and the deduced amino acid sequence was used to perform homology searches and multiple sequence alignments.
The mrns2-l mutant allele was amplified by PCR using the MD-2 yeast genomic DNA as a template, and cloned as a 1.1 kb BglII fragment into a general-purpose plasmid pTZ18R (Pharmacia),creating pTZ-MD2 (Table 3).
The entire mnzs2-1 sequences of three independent clones were determined. 2.2.7. PCR Amplification
Genomic DNA was isolated from the mms2-1 strain, WX17-4a. The
MMS2 open reading frame was amplified using the primers MMS2-4 and
MMS2-3 ( listed in Table 2). The PCR product contained the MMS2 coding sequence, 530 base pairs upstream and 300 base pairs downstream of the open reading frame; the sequence amplified allowed for analysis of the coding and regdatory regions of rn1ns2-1. The PCR mixture consisted of 10 pl of PCR buffer, 8 pl of 2.5 mM dNTP's, 5 pl of mrns2-4 and mms2-3 oligonucleotides at 5 w, 2 mM MgCI, 1.0 pl of Vent polymerase, 100 ng of genornic DNA, and sterile water; 100 pl total volume). Tl-,e PCR parameters were as follows: Step 1,94'C for 2 minutes; Step 2,94'C for 1 minute; Step 3,45'C for 1minute; Step 4,72'C for 1 minute; Step 5, retum to step 2 five times; Step 6,94'C for 1 minute; Step 7,
48'C for 1 minute; Step 8,72'C for 1 minute; Step 9, retum to step 6 twenty five times; Step 10, 72'C for 5 minutes; Step 11, hold indefinitely at 4'C. Table 2. Oligonucleotide sequences:
b Name Sequencea Position Source
M13-UP GTA AAA CGA CGG CCA GT N/A N/A
M13-RP CAG GAA ACA GCT ATG AC N/A N/A
MMS2-4 GGA CCC ACC ACT ATT CC MMS2
MMS2-3 GCA GAT CTA AGG GTT TAC t782 to t 799 MMS2
TCG AAT TCA AGT CCA CAC -1-1to t16 CAG CTA GGG AAT TCA TGT CCA CAC CAC; CTA G
AAC TGC AGC ATA ATA TCG GCT CGG C
a The oligonucleotide sequences are listed 593'. 2.2.8. Agarose Gel Electrophoresis
Unless otherwise stated, agarose gels were prepared at 0.75% agarose
(GibcoBLR) in 1 x TAE buffer (40 mM Tris-acetate, 2 mM Na,EDTA-îH,O). To
approximate the migration of the DNA, a sucrose dye (30% sucrose, 0.15%
bromophenol bIue, 0.1M EDTA, pH 8.0) was added to the samples. h DNA
digested with HindIII yields DNA bands of known size and these were used to
extrapolate the size of the DNA of interest. h DNA digested with HindIII yields
the following band sizes (in kilobases: 23.130,9.416,6.557,4.361,2.322,2.027,
and 0.564). Gels were electrophoresed using Fisher Biotech electrophoresis systems (Fisher Scientific, Pittsburgh, PA USA).
To visualize the DNA, the gels were stained in a diluted ethidium bromide solution (from a 10 mg/ml stock solution) for 3-5 minutes. The stained gels were viewed on the Fisher Biotech 312n.m Variable Intensity Transilluminator (FBTIV-614).
2.2.9. Isolation of DNA Fragments from Agarose Gel
The DNA isolation technique was adapted from Wang & Rossman (1994).
To isolate DNA fragments after digestion with restriction endonucleases, the digestion mixtures were mixed with sucrose loading buffer and loaded into the wells of a 0.6 % agarose gel. After electrophoresis was complete, the gel was visualized using a UV-illuminator; the DNA fragment of interest was identified and cut out of the gel using a sterile razor blade.
To remove the fragment from the agarose, the gel slice was filtered through a bed of saturated sephadex (Sephadex G-10; Pharmacia LKB) and silicon treated glass wool, supported witha 500 pl centrifuge tube with a hole in the bottom. The gel slice was placed upon the resin, and the tube was centrifuged witha 1.5 ml sterile centrifuge tube for 5-10 minutes, until the agarose was no longer visible. The resultant solution was treated with 0.1 ml of phenol and 0.1 ml of diloroform, and centrifuged for 3 minutes. The DNA contained in the top aqueous layer was precipitated with 2 volumes of ethanol.
2.2.10. Ligation
To ligate an isolated DNA fragment into a vector or plasmid, approximately 0.5 jig of DNA insert was combined with 0.2 pg of plasmid. 1 pl of T4 DNA ligase (Gibco BRL) and 4 pl of T4 DNA ligase buffer (5X stock solution) were added to the DNA, resulting in a total volume of 20 pl. Ligation was performed ovemight at 16°Cto promote the stable association of cohesive
DNA ends.
2.2.11. Probe Labelling
The DNA used for labelling was obtained by the isolation of DNA fragments from an agarose gel. The MMS2 probe was obtained by isolating the
1.1 kb fragment generated by a BglII digestion. The 0.8 kb SRS2 probe was generated by digesting the p14H construct with Pst1 and EcoRI. The 1.6 kb RAD6 probe was isolated after a BnrnHI-HindIII digestion of pB7, and the 5.4 kb
RAD18 probe from the ClaI generated fragment of pff11.156. Digestion of pJA6 with XbaI generated the 4.6 kb REV3 probe, and two RAD5 probes (4.2 kb and
3.85 kb) were generated after digesting pBM102 with HpnI. The DNA was labelled using the Random Primer Labelling kit from
GibcoBRL. As stated in the manufacturer's instructions, 25 ng of DNA is placed in a boiling water bath to denature the DNA, followed by immediate storage on ice. The DNA was combined with the random primers buffer mixture, dNTP's,
5OkCi of [32P]adCTP,and the Klenow fragment of E. di DNA polymerase 1, and incubated for more than 2 hous at 25'C. To stop the reaction, stop buffer was added, and the volume increased with water to allow for ethanol precipitation. The precipitated DNA was resuspended in 120 pl of water.
2.2.12. Southern Hybridization
The genomic DNA was digested with the appropriate restriction endonuclease, and the DNA fragments were separated on a 0.75% agarose gel.
The gel was treated in a solution of 0.25 M HCI for 10 minutes (depurination),
rinsed in distilled water, treated in 0.4 M NaOH/O.G M NaCl for 30 minutes
(denaturation), and then treated in 1.5 M NaC1/0.5 M Tris-HC1 (pH 7.5) for 30 minutes (neutralization). The DNA was transferred from the treated gel to
GeneScreen Plus membrane (Dupont) ovemight in a 10x SSC solution (20x SSC
stock: 3 M NaC1,0.3 M sodium citrate; pH 7.0). The DNA was crosslinked to the membrane the following day using a UV crosslinker (Fisher Sci. mode1 FB- UVXL-1000).
For Southem hybridization, the membrane was then placed into a hybridization bottle with 5 ml of pre-hybridization solution (2x SSC, 10%
dextran sulfate, 5x Denhardt's solution [50x stock: 10 g Ficoll,,, 10 g
polyvinylpyrrolidone, 10 g bovine serum albumin; water to 500 ml total
volume], 50% formamide, and 1%SDS), and incubated at 42°C in the hybridization oven for at least 2 hours. 50 pl of prepared carrier DNA (boiled then chilled, single stranded salmon sperm at 10 mg/ml) was added to the pre- hybridization solution, followed by adding 60 pl of probe. The membrane was incubated with the probe ovemight, and the following day was washed twice with cold wash (2x SSC, 0.1% SDS, room temperature) and twice with hot wash
(0.2~SSC, 0.1% SDS, 65'C) before the membrane was placed with x-ray film (Kodak, X-Omat film). The length of exposure depended on the strength of radioactive signal on the membrane at the end of the washes.
2.3. Plasmids and Plasmid Construction
The plasmids constructed in this study are listed in Table 3. A series of deletions was made within the insert of the library clone YCpM2 and the resulting plasmids were used to transforrn WX17-4a in order to map the MMS2 gene. The MMS2 gene was subcloned as a 3 kb XbnI-Hinm fragment from
YCpM2 into YCplac33 (Gietz & Sugino, 1988) to form YCp-MMS2 (Fig. l),and into pTZ18R to form pTZ-MMS2. A BamHI linker was inserted into the unique Ne01 site of pTZ-MMS2 to convert it to a BnmH site, which was then used to clone either the 1.8 kb BnnzHI fragment containing the LEU2 gene frorn YDp-L
(Berben et al., 1991) or the 1.2 kb BamHI fragment containging URA3 from YDp-
U (Berben et al., 1991), resulting in pmms2::LEU2 and pmmsZ::URAJ, respec tively. Plasmid pSCW-rad& was constructed by PCR amplification of the RAD6 gene with a mutation primer (RAD6-Ml), where an EcoRI site was created, followed by a mutated translation initiation codon (AAG instead of
ATG), and a 3' end specific primer (RAD6-2) to generate a BglII site 3' to the RAD6 translation stop codon. The EcoRI-BglII fragment was cloned into the EcoRI-BamHI sites of pSCW231 (Watkins et al., 1993) to place the rad6,,-, under
the control of an ADHl promoter. This mutation results in translation initiation of the rad6 gene at the second ATG encoding the 10th amino acid of Rad6
(Watkins et al., 1993). It has been detemined that the pSCW-radb,,,
transformants will produce AO-fold more Rad6 protein than wild type cells
(Watkins et al., 1993).
To clone the mms2-1 allele, the mutated gene was arnplified by PCR.
After the PCR amplification of the mrns2-1 allele the PCR reaction was treated with phenol and doroform, and ethanol precipitated. The resuspended DNA was digested with BglII. The digested insert DNA was ligated into BarnHI digested, phosphatase-treated pTZleR, creating pTZ-MD2. To study the effect of rnms2-1 overexpression, the mutated allele was cloned into the YCplac33 single copy vector and the YEplacl95 multi-copy vector. The mms2-l sequence was subcloned from pTZ-MD2, and ligated into BamHI digested, phosphatase- treated YCplac33 and YEplacl95 resulting in YCp-MD2 and YEp-MD2.
To measure the rate of P-galactosidase ubiquitination, we obtained the ubiquitin:P-galactosidase fusion constructs with Met, Arg, or Leu at the N- terminus of P-galactosidase (creating plasmids pUb23-M, pUb23-R or pUb23-L).
The consmicts were from Dr. Varshavsky (Caltech, Pasadena CA), and are described in Table 3.
Initially, hMMS2 and CROC1 were cloned by ligating BnmHI-Saci
fragments of hMMS2 and CROCl into pTZ18R digested with BnmHI and SRCI.
The ligation resulted in pTZ-hMMS2 and pTZ-CROCI. lzMMS2 and CROCl were also cloned into the pYES2.O yeast expression vector, placing bath genes under the galactose inducible promoter. The CROC1 and hMMS2 coding sequences were isolated after digesting pTZ-CROC1 and pTZ-hMMSZ with SstI and SnZI. pYES2.0 was digested with San and XhoI, then treated with phosphatase. Successful cloning yielded pYES-hMMS2 and pYESCROC1. For the mutagenesis assay, a 0.55 kb insert containhg hMMS2 was cloned into the YEp126 vector (Xiao et al., 1998). The pBlueScript-hMMS2 clone sent from
Human Genome Sciences Inc. was digested with EcoRI, generating the 0.55 kb fragment. The EcoRI sites are Iocated in the MCS of pBlueScript and in the 3' non-coding sequence of hMMS2. YEp126 was also digested with EcoN, and the hMMS2 fragment was ligated into the vector creating YEp-hMMS2. L, cll scn SRS2 in YCp50 F. Fabre RAD6 in YEp24 G. Simchen RAD18 in YCp5O F. Fabre
RE V3 in YCp50 A, Morrison RAD5 in YEp24 L. Prakash 1.1 kb containing nims2-2 in pTZl8R This study. rms2-2 in YCplac33 URA3 This study. mms2-1 in YEplacl95 URA3 This study. ADH4:: URA3:: (TG .& Gottschling, et al., 1990.
Ub::P-galactosidase fusions 2m URA3 A. Varshavsky
BamHI-Sac1 fragment of IiMMS2 This study. in pTZ18R
BamHI-Sac1 fragment of CROC1 This study. in pTZl8R hMMS2 in YEp-URA3, pGAL1-tCYC1 This study. CHAPTER THREE: RESULTS
3.1. Molecular Cloning Of MMS2
3.1.1. Isolation Of Clone From Genomic Library The mms2-l mutant was originally isolated by Prakash and Prakash
(1977) for its enhanced sensitivity to MMS. The MMSZ gene was cloned by screening a single-copy yeast genomic library for the functional complementation of the mms2-1 MMS-sensitive phenotype. The library constructed by Rose et al. (1987) utilizes the centromeric YCp50 plasmid ensuring low copy number and mitotic stability. This library was chosen for two reasons. Firstly, if MMSZ were toxic at high levels a multi-copy library would select against MMS2 isolation. Secondly, using a multi-copy library may select for false positives, including extragenic suppressors of mrns2-1.
Over 10,000 transformants were screened to obtain a clone able to suppress the MMS sensitive phenotype of rnms2-2. The plasmid responsible for the complementation was isolated from the WX17-4a mutant and was transformed into DH5a to amplify the plasmid DNA. The initial MMS2 clone,
YCpMZ, contains a 10 kb insert. To confirm that the YCpM2 clone was responsible for the MMS resistance, a co-segrega tion test was performed (Section 2.1.6.). The WX17-4a colonies that were unable to grow on SD-Ura plates were sensitive to MMS. The WX17-4a colonies that were Ura' were resistant to MMS. These results confirmed that the YCpM2 plasmid was responsible for complementing the MMS sensitive phenotype of WX17-4a.
3.1.2. Mapping The YCpM2 Clone
As previously stated, the YCpM2 clone contains a 10 kb insert. To determine which ORF within YCpM2 was responsible for complementing ntrns2-
1, various deletions were constructed and tested for their ability to complement the MMS sensitivity of WX17-4a. A crude restriction map was constructed by digesting YCpM2 with various restriction enzymes. After deleting approximately 5 kb of sequence (SnlI-PuuII double digestion) the clone was still able to complement the MMS sensitive phenotype of WX17-4a. The remaining insert was subcloned into a YCplac33 centromeric vector (URA3 marker), creating YCp-MMSZ. Sections of YCp-MMS2 were deleted to identify the region of insert that contained MMS2, to avoid sequencing extraneous insert DNA. Since the XbnI deletion construct was able to complement the MMS sensitive phenotype of mrns24, it was used as the template.
The sequencing results were compared ro known sequences in the non- reduntant GenBank CDS database using the BLAST program (http://www.ncbi.nlm.nih.gov/BLAST/) to determine if the sequencr had been previously documented (Alhchul et al., 1990). The insert sequence obtained with YCp-MMS2 XbnIA (M13-RP primer) was identical to MADI. The other strand of YCp-MMS2 XbaU (M13-UP primer) was sequenced, yielding a sequence identical to SNRIO. Both SNRlO and MADZ reside on Chromosome 7. To discern which loci was responsible for complementation, YCp-MMS2 was digested to create deletions in either MADl or SNRZO. The YCp-MMS2 XbaIA construct was digested with BstEII and EcoN (from the MCS of the vector, YCp33) to remove part of the SNRIO sequence. This remaining fragment was still able to complement the MMS sensitivity of mms2-1 (Fig. 4). Removal of a
Sac1 fragment (disrupting MADI) did not affect complementation of mms2-1 (Fig. 4). This indicated that neither MADl nor SNRIO were allelic to mrns2-1.
A putative ORF was identified between the MAD1 and SNRlO loci after further sequencing. To distinguish whether the ORF was MMSZ, the gene was disrupted by a SnlI-NcoI (SaII from the MCS) deletion and a EcoN-NcoI deletion.
Neither deletion construct was able to complement the MMS sensitivity of mms2-1, indicating that the putative ORF was indeed MMS2. Thus, a combination of deletion mapping and sequencing was able to determine that the
MMS2 gene (GenBank accession number U66724) is located on Chromosome 7 and resides between MADl and SNRIO (Fig. 4).
3.2. Initial Analysis Of iW2
3.2.1.lMlMS2 1s Allelic To mns2-l
To confirm that the ORF within YCp-MMS2 was MMSZ, the ORF was disrupted using LEU2 and URA3 marker genes (Section 2.2.13.). Two pieces of evidence strongly suggest that the cloned MMS2 gene is allelic to mms2-1. First, the cloned MMS2 was used to disrupt the chromosomal MMS2 gene and the mutant became sensitive to MMS (Fig. 7); this mms2 mutation was unable to complement the rnrns2-1 mutation in a diploid. Secondly, the mms2-1 allele from MD-2 was cloned by PCR amplification and a C-to-T transition at nucleotide 303, which results in a single amino-acid substitution P73L (Section 3.3.1.; Fig.5), Sac l Xbal Ncol Bglll 1 Hindlll
L r MMS MADI MMSB SNRlO
Figure 4: Mapping and disruption of the MMSZ gene. A subclone containing the 3 kb BarnH1-PuuII fragment (YCp-MMSZ) enables (+) the rnrns2-1 mutant to grow on YPD plates containing 0.4% MMS. Further deletions to the NcoI site from either end abolished (-) the MMS2 function. Either a URA3 or a LEU2 fragment was inserted at the NcoI site to construct the mrnsZ::URA3 and mmsZ::LEU2 disruption cassettes. was identified. Three clones were sequenced, eadi obtained from an independent PCR reaction, and each displaying the identical nucleotide substitution.
3.2.2. Phenotypes Of mms2
In contrast to the mms2-1 mutant, the mms2 disruption mutant is sensitive to MMS in bath plate and liquid assays (Fig. 7). In addition, compared to an isogenic wild type strain, the nirns2A mutant is more sensitive to W (Fig. 9). Thus the role of the MMSZ gene appears not to be limited to the protection of cells from DNA methylation damage. Our mms2 disruption mutant is likely a complete loss-of-function mutant since it was as sensitive to MMS and W as an mms2A strain T43 where the entire MMS2 gene is deleted.
3.2.3. Sequence Analysis Of MMSZ
3.2.3.1. Analysis Of Nucleotide Sequence
MMSZ encodes a predicted 137 amino acid, 15.5 kDa protein with a single intron (Fig. 5). The splicing sequences located within S. cerevisiae introns are highly conserved (Rymond & Roshbash, 1992). We confirmed that the first exon encoding four amino acids is required for MMS2 function, since MMS2 clones using the second ATG codon, at nt 218, as the translational start were unable to complement the mms2-1 mutation, whereas an intron-less MMS2 clone is able to complement the MMSsensitive phenotype of mms2-2. 3.2.3.2. Analysis Of Amino Acid Sequence
The C-terminal two thirds of the deduced Mms2 shares significant similarity with almost all known Ubc proteins, with P values ranging from IO-" to 10" as determined using the BLASTp program (BLOSUM62 matrix) (Altschul et al., 1990). The middle third of Mms2 shares up to 40% identity and 60% similarity with some Ubc proteins (Figo), suggesting an evolutionary conservation. Like Ubcs, the deduced Mms2 is rich in proline residues (12/137), particularly in the middle one-third region (8/41), indicating that it rnay form a globular protein. Despite the high overall degree of homology between Mms2 and Ubcs, Mms2 lacks a critical consensus sequence which al1 Ubc proteins display around their active Cys residue (Fig. 6). The absence of this consensus sequence suggests that Mms2 does not possess direct ubiquitin conjugating ability. Indeed, partially purified Mms2 from an E. coli expression system was unable to bind Ub and lacked E2 activity (V. Chau, Wayne State University, persona1 communication). Mms2 has even stronger homology (P = IO-") with
Crocl, a recently identified transactivator of the c-fos enhancer (Rothofsky & Lin,
1997) which also lackç E2 activity (Sancho et al., 1998). The two proteins share
50% amino acid identity throughout the entire length of Mms2 (Fig. 25).
i3giII AGATCTCTAGTCCCAArnCA <--MW1 GPGAA'PCGIY:CACIY]AATGCAM'GAAGAAACAmAAn;cCGCTCIYlACA'PCCA~AmAmAwACCTCTCGAmAAGATGGAAA AAAmAGGAAmAmCTAGGGTAACACGCTATCA CTACACAATCrCTGCrCATTACATITCGAn;TCGTGGn;AAA'ITCTTAmTATATGCAACGTAGAAGAAAGCA~ACACARAA A~utatuttattataagttttgaaagggatttcaacata~aattcttgaaatgtaccctcactgtcctattgtaa MSKV at~~TTTT~AMAaMlT~TCATd~A~~T PRNFRLLEELEKGEKGFGPESCSYGLAD ~TAmiAccATCUICCAMMOAACGOCACTAmAOOOCCCCCPCATACrPMTCATOAAAACA[UATCTAmCCI"CPCPA'rA SDDITMTKWNGTILGPPHSNHENRIYSLSI t OATPCIPOOCCCAAkCT~~TAAATCTAccAmcQTcM~~ DCGPNYPDSPPKVTFISKINI1PCVNPTTGE NCOI E~irXTT~~~ VQTDFHTLRDWKRAYTMETLLLDLRKEMAT ~~iATCTA.mlCl'AAACAGTA?TIY;CAûCA1TCCAAûCC PANKKLRQPKEGETF* AmTATATACAAGCAATAAGACATAGATATCmAmAGGTAGTAAGTITAATTACI*ITCATAATATGAATTTACTCACCCCAm 1 <--SNRIO CCGAI;AAGTTACI'CGK;CAA?TCACTACGATAA'IY;A?Y;TATATAGTIY;TATTATCAA'IY:CTK;CAACGGTCCTCGTGCAATIYlACTACGA Bgl II TAA'IY;An=TATATAGI*K.TATTAWAATCCTKiCAACGGTCC1Y:ATCCGGGCACACCAACGTAAACCCITAGATCT Figure 6: Çequence alignment of Mms2 and other Ubcs in S. cerevisiae. Highlighted areas indicate conserved arnino acid residues. The asterisk denotes the conserved cysteine residue that is required for Ubc activity. 3.3. The mms2-l Allele
3.3.1. Cloning And Sequencing Of mrns2-1
To determine the genetic alterations in the mms2-2 allele, primers
designed from the cloned MMS2 (MMS2-4 and MMS2-3; Table 2) were used to amplify genomic DNA hom WX17-4a by PCR. DNA from three separate PCR reactions was sequenced to confirm that any genetic alteration in mms2-1 allele
was not due to PCR-mediaied mutation. Each sample shared the same sequence
alteration; namely a C+T transition at nucleotide 303, causing a P73L
substitution (Fig. 5). No further mutations were found in the ORF or the
promoter region.
3.3.2. rnms2-1: A Partial Loss-Of-Function Mutation
The original MD-2 strain (containhg the mrns2-1 allele) was characterized
as sensitive to MMS but not to UV-induced killing (Prakash & Prakash, 1977).
We found that the mms2-1 mutant is sensitive to MMS by a plate assay, but not
significantly sensitive in a liquid assay (Fig. 7). The mnrs2-1 mutation results in
only partial loss of function, since a single-copy of mms2-1 (in centromeric plasmid) was unable to rescue the mms2 mutant from killing by MMS, yet mms2-
1 in a multicopy plasmid (2p based plasmid) was able to complement the mms2
mutant (Fig. 8). It should be noted that P73 mutated in rnms2-1 is consenred in
al1 the corresponding Ubcs (Fig.6) as well as in Croc1 (Fig. 25), and that it is
adjacent to another highly conserved proline residue, which, when mutated
(e.g., P64S in Rad6 and P71S in UbcJ), results in a temperature-sensitive
phenotype (Ellison et al., 1991). O 2 0 4 0 6 0 8 O Minutes in 0.3% MMS
Figure 7: Cornparision of the MMS sensitivities of the mms2 null mutant and the nims2-1 allele. The mnis2 null mutant ( .) shows increased sensitivity to MMS in liquid culture, compared to the corresponding wild type 747 ( ). The MD-
2 strain ( () ), containing the mnis2-1 allele, and its isogenic wild type strain
B635 ( O ) did not display the same difference in MMS sensitivity as seen between 747 and SBU (mms2 null mutant). The results are an average of three independent experiments. Strain
Figure 8: Over-expression of the mms2-l allele complements the MMS sensitive phenotype of the mms2 nul1 mutant. The growth pattern displayed is of cells on
0.04% MMS gradient plates. Strain 1, FY86 (wild type); strain 2, M86Am2; strain
3, FY86Am2 transformed with YCp50 (vector control); strain 4, FY86Am2 with
YCp-MMS2; strain 5, FY86Am2 with YEp-MMSZ; strain 6, FY86Atn2 with YCp-
MD2; strain 7, M86Am2 with YEp-MD2. 3.4. IM2MS2: Genetic Interactions
3.4.1. MMS2 Belongs to the RAD6 Epistasis Group
nie relative levei of MMS and UV sensitivity of the mms2 nul1 mutant, as well as the Ubc-like sequence of the Mms2 protein, led us to speculate that MMS2 may function in the RAD6 pathway. To test this hypothesis, we created mms2 rad6A and mms2 radl8A double mutants and found that, as expected, the double mutants were no more sensitive to UV (Fig. 9A and 98) and MMS (Fig.
10A and 10B) than the respective rad single mutants. Thus rad6 and rad78 are epistatic to mnis2, indicating that MMS2 belongs to the RAD6 group.
3.4.2. MMSZ 1s Not A Member Of NER Or Recombination Repair
We also performed epistatic analysis of mms2 with nucleotide excision repair (rnd4A) and recombination repair (rad50A) mutations. The mmd rad4A
(Fig. 9C) and mms2 rnd50A (Fig. 9D) double mutants were found to be more sensitive to W than either of the corresponding single mutants and the killing effects appeared to be additive. Similar results were obtained when rad4 mms2 and rnd50 mms2 cells were exposed to MMS (Fig. 10C and 10D). T'usthe MMSZ gene does not belong to the RAD3 or RAD52 epistasis groups; it is specific for the RAD6 pathway. Figure 9: Assessing the genetic interactions of mms2 in response to UV- irradiation. DBY747 ( ) wild type and mms2 ( .) cells. The percent sumival of mms2 rnd6, mms2 radl8, mms2 rad4 and mms2 rad50 double mutants was calculated. The sensitivity of the mms2 rad6 double mutant ( )(A) was equally as sensitive as the rad6 single mutant ( O ). Similarily, the mrns2 rad18 double mutant ( ) displayed the same level of UV sensitivity as the rad18 single mutant ( O ) (B). These results place MMS2 within the RAD6 epistasis group. After UV irradiation both the mms2 rad4 ( ) (C)and mms2 rad50 ( )
(D) double mutants display a UV sensitivity greater than their corresponding single mutants, rad4 ( O ) and rad50 ( O ) Hence, mms2 does not belong to the rad4 pathway (NER) or the recombination repair (rad50) pathways. 0.0001 I 1 1 I O 50 100 150 2( O 50 100 150 2( UV Dose (Jlrn 2) UV Dose (Jlm 2)
UV Dose (Jlm 2) UV Dose (Jlm 2) Figure 10: Assessing the genetic interactions of mms2 in response to MMS. The percent sumival of mm2 rad6, mms2 rad1 8, mms2 rad4 and mms2 rnd.50 double mutants( O ) was calculated. The mms2 mutant ( ) strain and DBY747 wild type ( ) were included as well. The sensitivity of the mms2 rad6 double mutant (A) was equally as sensitive as the rad6 ( .) single mutant. Similarily, the mms2 rnd18 double mutant displayed the same level of MMS sensitivity as the rad18 ( fl ) single mutant (B). These results place MMS2 within the RAD6 epistaçis group. After treatment with MMS, both the mms2 rad4 (C)and mms2 rad50 (D) double mutants display a MMS sensitivity greater than their corresponding single mutants, rad4 ( ) and rad50 ( .). Hence, mms2 does not belong to the rad4 pathway (NER)or the recombination repair (rad50) pathways. O 2 0 40 60 8 0 O 2 0 4 0 6 0 8 0
Minutes in 0.1% MMS Minutes in 0.1% MMS
O 20 40 60 8 O 2 0 4 0 6 0
Minutes in 0.3% MMS Minutes in 0.15% MMS 3.4.3. MMS2 and REV3 Mutagenesis
It is predicted that another gap-filling activity exists within the RAD6 pathway that is independent of REV3-dependent mutagenesis. These predictions are based upon two observations. Firstly, unlike rad6 and rad18 mutants, rev3 mutants are only moderately sensitive to UV and MMS. Secondly, reu3 mutants do not display the rad6 or rad18 defect in gap-filling, the hallmark of PRR mutants. 1 wanted to discern how MMS2 was involved in PRR. Similar to rev3A mutants, nrms2A cells are significicantly less sensitive to MMS and UV than the rod6A and rndl8A mutants. The moderate MMS and UV sensitivity of mms2 could be explained by two different hypotheses. Firstly, hiIMS2 may provide a PRR function that can be replaced by another protein, reducing the dependency for Mms2 after DNA damage. Secondly, MMS2 may be a member of a sub-pathway of the greater RAD6 pathway, as seen in the example of RE V3.
The first question 1wanted to address was whether mms2 behaved similarily to rad6 and rd18 regarding spontaneous mutagenesis. The spontaneous mutation rates of mms2 were determined using the trpl-289 reversion assay. The nznzs2 single mutant increased the spontaneous mutation rate by >30 fold at the trpl-289 allele (Table 4). The elevated spontaneous mutation rate observed in mms2 was quite dramatic and opposed the idea that mms2 may be involved in the mutagenic response. To better address the mutagenic relationship between mms2 and reri3, the spontaneous mutation rate of the rev3A single mutant and the mms2 rev3A double mutant was measured.
The spontaneous mutator phenotype of ntrns2 is completely dependent on the functional REV3 gene, since the spontaneous mutation rate of the mms2 rm3A double mutant is as low as the rev3A single mutant (Table 4). The spontaneous mutagenesis results indicated that MMS2 does not participate in mutagenesis. To investigate the role of MMS2 within the RAD6 pathway, we measured UV-and MMSinduced killing of the mms2 rev3A double mutant. To Our surprise, while each of the mm2 and rev3A single mutants were only moderately sensitive to the DNA damaging agents, the mms2 rev3 double mutant was extremely sensitive to both UV (Fig. 11A) and MMS (Fig. 11B). The effect of the two mutations was clearly synergistic, since the fractions of mms2 rev3A surviving after the highest dose of UV treatment (50 J/m2)or 40 minute treatment with 0.l0/0 MMS were 1,000 fold lower than expected if the mms2 and rm3A effects were simply additive (Fig. 11A and 11B). This result led us to speculate that MMS2 represents a repair pathway that is an alternative to REV3 mutagenesis.
The rev3 mutant was not isolated for drug sensitivity, but was isolated for a defect in UV-induced mutagenesis. To confirm that MMS2 was not a member of the REV3 mutagenic pathway, we compared the UV-induced mutation rates of mms2 and rev3. We predicted that if MMS2 is involved in the
REV3 mutagenic pathway, mms2 cells should display a similar decrease in mutagenesis. As seen in the spontaneous mutagenesis assay the UV-induced reversion of the hetrpl-289 allele is also elevated in the mms2 mutant (Table 5), unlike rev3 mutants which are completely defective in DNA damage-induced mutagenesis. This result is consistent with a role for MMS2 in PRR apart from REV3 mutagenesis. I 1 I 0.01 1 I 1 1 0.001 1 I O 50 100 150 200 O 2 0 4 0 6 0 8 O UV Dose (Jlm 2) Minutes in 0.1% MMS
Figure 11: mms2 is synergistic to reu3 with respect to both UV (A) and MMS (B) sensitivity. ( ), DBY747 (wt); ( .), SBU (rnms2);( O ), WXY9382 (rev3A); ( ), SBUr3L (reu3A nzms2). The results are an average of two independent experiments.
3.4.4. rad6Al-9 is Epistatic to mrns2A. Some of the mms2 mutant phenotypes, including the intermediate level of
W sensitivity and increased mutagenesis, are remarkably similar to the rad6 mutant with an N-terminal deletion, rad6,,-,, described by the Prakash laboratories (Watkins et al., 1993). We reasoned that the rnd6,,-, mutation may also affect the error-free PRR but not the mutagenesis pathway. A critical experiment would be to see if rnd6,,, and mms2 mutations show an epistatic interaction. We constructed a rad6,, allele in a pSCW231 plasmid as previously described (Watkins et al., 1993). The resulting plasmid, pSCW-rad6,,.,, was used to transform rnd6A and mms2 rnd6A mutants to create rad6,,, and mms2 rad6,,, strains, respectively. While rad6,,, partially rescued rad6 sensitivity, epistatic analysis showed that the mms2 rad6,,, double mutant was as sensitive to killing by both UV (Fig. 12A) and MMS (Fig. 12B) as the rnd6,,, single mutant. Thus, both rad6A and the rnd6,,-, deletion are epistatic to rnms2A with respect to UV and MMS sensitivity . 0.0001 1 1 I 1 I O 50 100 150 200 UV Dose (Jlm 2) Minutes in 0.3% MMS
Figure 12. rad6,, is epistatic to rnrns2A with respect to both UV (A)and MMS (B) sensitivity. ( ), DBY747 (wt); ( .), SBU (mnzs2);( O ) ,WXY9376/pSCW- rad6,,, (rnd6,,-,); ( ), SBU6L /pSC W-rad6,,, (mms2 rnd6,,,). The results are an average of two independent experiments. 3.4.5. Epistatic Analysis Within Error-free PRIX
Other groups have reported the isolation of mutants suspected of a role in error-free PRR, induding poZ30-46, rad5 and rnd30. In each case, it is not disputed that these new PRR members are hypostatic to rad6, but it must be established how they genetically interact with one another.
As mentioned in section 1.6.2.4.2., the RAD5 gene is implicated in the error-free PRR pathway. Epistasis analysis was performed using rad5, mrns2, and rad5 mms2 mutants in the DBY747 strain. The sensitivities of these mutants was analyzed using UV and MMS as the source of DNA damage. The UV and
MMSinduced killing was more severe in the double mutant than in the corresponding single mutants (Fig. 13A and 138). The additive sensitivity observed in the double mutant compared to the single mutants indicates that
RAD5 and MMS2 do not act within the same error-free PRR pathway.
To study the possible roles of the essential gene POL30, alanine scanning mutagenesis was used to aeate various POU0 alleles (Ayyagan et al., 1995).
One specific mutant, PY39-46, appeared to have a defect in DNA damage repair
(Ayyagari et al., 1995); epistasis analysis indicated that the poZ30-46 allele was responsible for the error-free PRR defect in PY39-46 (Torres-Ramos et al., 1996).
1 wanted to examine how the mms2 and ~0130-46alleles interacted genetically.
We obtained the PY39-0(wild type) and PY39-46 (pol30-46) strains from Dr. P.
Burgers (Washington University School of Medicine, St. Louis MS) (Table 1).
We cowtntcted the PY39mms2 poZ30-46 and PY39-mms2 mutants so that the mms2 mutation is present in an isogenic background. 1 assessed the W and
MMSinduced killing of mrns2, pol30-46 and the double mutant. In both UV and
MMS treated cells the double mutant was more sensitive to the DNA damage than the corresponding single mutants (Fig. 14A and 148)- The phenotypes of
the individual poL30-46 and mms2 mutants indicate they share a role in RAD6-
dependent error-free PRR, but the additive effect of the double mutation
indicates the genes do not hction within one specific error-free PRR process.
McDonald et al. (1997) isolated a yeast DinB and UmuC homologue,
RAD30; epistasis analysis and mutagenesis studies implicated RAD30 in an
error-free PM. Johnson et al. (1999b) discovered that Rad30 was a polymerase
that could synthesize DNA past sites of DNA darnage, providing the first
biochemical mechanism for error-free PRR. To establish whether Mms2
functioned with Rad30 in hanslesional bypass, epistasis analysis was performed
with rad30, mms2, and rad30 mms2 mutants. As observed in po130-46 mms2 and
rad5 mms2 epistasis analysis, the rad30 mms2 double mutant was more sensitive
to W and MMS than either mms2 or rad30single mutants (Fig. 15A and 158).
Evaluating how MMS2 interacts with the other members of error-free PRR has provided more questions than answers about the darnage tolerance
mechanisms within the pathway. Epistasis analysis suggests that Mms2
functions within a separate pathway from Rad5, Rad30 and Po130, making it
more difficult to resolve the mechanism of DNA darnage tolerance. A possible explanation for these genetic interpretations may be that these genes are
involved in multiple DNA repair pathways. Involvement of an enzyme in two different pathways, both responsible for the repair of a comrnon lesion, could show additive sensitivity to the drug causing the cornmon lesion. For example,
Po16 and Po130 are necessary for DNA replication, but have secondary functions in more than one DNA repair pathway. O 25 50 75 100 125 O 2 0 40 60 8 0 UV Dose (Jlmz) Minutes in 0.1% MMS
Figure 13: Epistasis analysis between rad5 and mms2 within error-hee PRR. rad5 ( O ), mms2 ( ), and rad5 mms2 ( ) nul1 mutants were created in a
DBY747 ( a) background. Al1 four strains were treated with UV (doses ranging from 20-100 J/mZ),and 0.1% MMS for up to 60 minutes. The results are an average of three independent experiments. O 20 40 60 80 100 120 O 20 40 60 80 UV Dose (J/m2) Minutes in 0.1% MMS
Figure 14: Epistasis analysis between mms2 and poZ30-46. A mms2 nul1 mutation
was created in the PY36-46 strain (pol30-46 allele) to create the po130-46 mms2 double mutant ( ) Compared to the mms2 ( .) and pol30-46 ( O ) single mutants, the double mutant shows a greater sensitivity to both (A) UV and (B)
MMS. The PY39 strain ( ) was included as the isogenic wild type control.
The results are an average of three independent experiments. O 25 50 75 100 125 O 20 4 O 6 0 8 0 UV Dose (J/m*) Minutes in 0.30h MMS
Figure 15: Assessing the involvement of MM52 in translesional bypass. The
T43 strain ( O ) bearing the mms2 nul1 mutation, was used to create the T43 rad30 ( ) double mutant. BY448 ( a), the isogenic wild type strain, was used to create the rad30 ( .) single mutant. Survival of the double mutant after treahnent with both (A) W and (B) MMS was less than either single mutant. The results are an average of three independent experiments. 3.5. Mms2 Accessory Protein To Rad6 Or An Enzyme Exclusive To PRR?
The Rad6 ubiquitin-conjugating enzyme is involved in many diverse
cellular processes (see Section 1.6.2.2.). How the ce11 focuses the activity of Rad6
specifically for PRR rernains unknown. MMS2 shares with RAD6: i) an
involvement in PRR, and ii) structurally similarity (Rad6 is an E2, and Mms2 is
related to EZfs). Epistasis anaiysis indicated that MMS2 did not participate
specifically in RADS-,RAD30 - or POL30 -dependent error-free PRR, so 1
wanted to address whether Mms2 acts as an accessory protein for Rad6-
mediated activities.
3.5.1. Sporulation
Homozygous rad6 diploids display a defect in sponilation. Depending
on the strain background, rnd6/rad6 diploids have sponilation levels 0.0 to 4.0 %
of the isogenic wild type strain (Morrison et al., 1988). The sporulation activity
of Rad6 requires its ubiquitin-conjugating adivity, as well as the aàdic carboxyl-
terminus tail (Momson et al., 1988). To address whether mms2 homozygous
diploids share a sporulation defect with rad6 diploids, wild type and mmd
haploid strains were crossed; M86 (MATa)was crossed to XS-803-3A (MATa)
and FY86 mms2::LEU2 (MATa)crossed with XS-803-3A mms2::LEU2 (MATa).
The cells were crossed on SC-histidine plates to select for diploids and against either haploid from growing. The diploids were then induced to undergo sporulation; selected colonies were diluted, and the number of tetrads were
quantitated using a hemocytometer.
Two separate sponilation experiments were conducted, with 4 individual diploid isolates tested each tirne. Formation of tetrads was used as an indication of sponilation. Table 6 shows the percent sporulation based on the XS803-
3A/FY86 (wild type) values. The sarne experiment was conducted using rad6 diploids cells (from the sarne genetic background FY86 rad6 aossed with XS
803-3A rado), but the cells did not produce any tetrads.
The mms2 diploids induced to sporulate produced tetrads, but the percent of tetrads formed was reduced compared to wild type diploids. In
Experiment 1, mm2 diploids formed tehads at levels 37.8%that of wild type diploids (Table 6),with a standard deviation of 22.7. Experiment 2 produced sirnilar results; mms2 diploids were able to fom tetrads at 51.5% (standard deviation 21.5) that of correspondkg wild type diploids (Table 6).
The results from this experiment indicate that mms2 is not essential for spomlation. The reduction of tetrad formation in mms2 cells does not necessarily reflect a role in sponilation, but may indicate a secondary defect created by unresolved spontaneous DNA damage.
3.5.2. Telomeric silencing
Telomeric silencing is the phenornenon obsewed in S. cermisiae where
genes placed within or adjacent to telomeric sequence become transcnptionally
silent, which is also referred to as telornere position effed (Gottschling et al.,
1990). Many genes have been identified that are specifically involved in the
tight regulation of telomeric DNA (Apanao et al., 1991). Recently, rad6 mutants
have been shown to possess a defect in telomeric silencing (Huang et al., 1997).
A method was established by D. Gottsdiling to monitor the level of telomeric
silencing. Placement of the URA3 gene into the telomere of S. cermisiae provides
a method to detect its transcriptional activity (Gottschling et al., 1990). The gene
product of URA3 metabolizes 5-fluoro-orotic aad (FOA) into a toxic intermediate, so that any cells expressing URA3 are unable to survive on an
FOA plate (Boeke et al., 1987). By observing the level of growth on FOA plates
mutants containing the URA3 telomeric marker can be evaluated for a defect in
telomeric silencing. The URA3 gene was integrated into the telomere of chromosome Vn of
DBY747mms2. 10X senal dilutions were made, and the dilutions were replica- plated ont0 YPD and FOA plates. Identical growth was seen for the DBY747mms2::LEU2 TEL::URA3 mutant on YPD and FOA plates, indicating no loss of telomeric silencing (Fig. 16). As a negative control, yeast strain tlcl.8D was included; this strain does not exhibit the telomere position effed, and the growth of these mutants is inhibited on FOA plates (Fig. 16). These results indicate that Mms2 does not affect Rad6-mediated telomeric silencing. III
YPD FOA
Figure 16: The mms2 mutation does not alter telomere silencing. The URA3
gene was incorporated into the telomere of the DBY747mmsîA skain (1), the
YPH925 strain (II) and the tld-hurnan.8D strain (III). YPH925 was used as the
positive conbol for telomenc silencing. tld-human.8D had been engineered to
express human telomenc sequence, making it defective in telomeric silencing.
tld-human.8D is the negative control. 10X serial dilutions of each strain were
replica-plated ont0 YPD and FOA plates. Growth on FOA indicates that the
URA3 locus is not expressed, and the telomeres are transcriptionally silent. 3-53. Protein Degradation via the N-end Rule
The N-end rule of protein degradation is the process in which the amino terminal amino aad determines the protein half-life in vivo. The recognition of the protein target by the E2 and E3 enzymes relies on two key amino acid signals contained within the target protein, a signal referred to as the N-degron
(Varshavsky, 1997). The N-degron consists of a destabilizing N-terminal residue and an intemal lysine(s) residue (Varshavsky, 1997). In S. cereuisiae, a hierarchy exists as to which amino aads at the amino (N-) terminus of proteins are responsible for rapid ubiquitination. Proteins with Asn, Gln, Asp, Glu, Leu,
Phe, Tyr, ne, or Trp at their N-terminus are further modifi ed by the enzymatic addition of an Arg residue. N-terminal Arg is a primary destabilizing residue that is immediately bound by the E3 enzyme of the ubiquitin pathway, promoting rapid degradation (Varshavsky, 1997). Methionine is a stabilizing residue. and is not recognized by the N-degron.
Rad6 (Ubc2) is required for protein degradation mediated by the N-end de(Dohmen et al., 1991). The assay used to monitor protein degradation via the N-end rule introduces a Ub-X-pgal fusion protein, where X represents variations in the N-terminal residue of the Pgal protein (Badunair et al., 1986).
Eukaryotes possess Ub-specific proteases that recognize and cleave the Ub moiety off the hybrid molecule, leaving a X-pgal protein. We obtained three Ub-
X-pgal constructs to examine the involvement of Mrns2 in the N-end mle: Ub-
Met-pgal, Ub-Arg-pgal, and Ub-Leu-pgal. Methionine at the N-terminus is a stabilizing residue, conferring a longer in uivo half life to proteins. When present at the N-terminus, Arg and Leu are destabilizing residues, quickly targeting proteins for ubiquitination. Using the Pgal assay, cells proficient Met Leu AW
Figure 17: The ~galactosidaseactivity, as a measure of ubiquitination via the
N-end de. Each set of three colurnns represents strains containhg
galadosidase with Met Leu or &g residues atthe -terminus,------respectively. ----- The stability of each of the above consmicts was tested in FY86 (wild type), FY86 mms2A and FY86 rad6A, and the average of four experirnents is shown. The standard deviation is given by the error-bars on the graph. The values for FY86 mms2A and wild type are significantly different (P < 0.05) for al1 arnino acids tested using the matched-pairs t test. in the N-end rule ubiquitination should have an immediate reduction of Pgal adivity when transformed with Arg-pgal or Leu-pgal plasrnids. Dohmen et al.
(1991) have shown that rad6 cells have stabilized Arg-Bgal and Leu-pgal protein to a level comparable to the stable Met-pgal control, indicating that Rad6 is the
E2 enzyme involved in N-end rule mediated ubiquitination.
We transformed the three X-pgal plasmids into FY86 (wild type), FY86 mms2, and WB6 rnd6 backgrounds. Cells were prepared as in Section 2.1.12. As represented in Figure 17, the Met-pgal protein is stable in all three strains. After the 30 minute incubation, the kg-pgal and Leu-pgal proteins had been almost completely degraded in FY86, expressing a relative &al activity of 7.88% and
0.89%, respectively (Fig. 17). In the mms2 nul1 mutant the Pgal adivities of Leu-
Pgal and Arg-&al were 29.5% and 3.54% of Met-pgal, respectively (Fig. 17). rad6 mutants displayed a defect in protein degradation, indicated by the increased stability of Pgal with Leu and Arg at the N-terminus; Leu-&al and hg-Bgal constructs expressed Pgal activity that was 89.2% and 79.5% that of
Met-pgal (Fig. 17). It appears as though the FY86 mms2 mutant does display a modest increase in Leu-pgal and Arg-pgal stability.
3.5.4. Associated Functions Of Mms2.
rnms2A mutants show a moderate difference, compared to isogenic wild type strains, in sporulation and in protein degradation via the N-end rule.
These moderate phenotypes make it difficult to assess how MA4S2 affects these processes, since the results are intermediate between the positive and negative controls. There are some possible explmations as to how Mms2 may partially affect spodation and protein degradation. Mms2 may function with Rad6 in sporulation and protein degradation, but its activity may be substituted in the event that Mms2 is not available. Altematively, Mms2 may also represent a
Rad6-independent mechanism. In this situation, the activity provided by Rad6 is preferred over the activity of Mms2. A more detailed discussion is made in
Section 4.3.
3.6. SRSS: Involvement in Error-free PRR
3.6.1. Epistasis between srs2 and mms2 The error-free PRR pathway recently defined by MMS2 was based on the epistasis between rad6 mms2 and rad18 rnrns2. To address the relationship between MMS2 and the suppressor of rad6 and radZ8, epistasis analysis was performed between srs2 and mms2. The experimental goal was to determine the extent of suppression within the PRR pathway. As seen in Figure 18A, srs2 cells display a greater MMS sensitivity compared to the mms2 single mutant. The srs2 mms2 double mutant is as sensitive to UV as the srs2 single mutant, indicating the epistasis between srs2 and mms2. Similar results were obtained when cells were treated with MMS (Fig. 18B).
The data obtained with srs2 rad6 or srs2 rad18 could be explained by two models: i) Srs2 acts to promote Rad6-dependent PRR, preventing the channehg of lesions through recombination repair, or ii) Srs2 activity provides a pivotai function within PRR. Epistasis analysis places srs2 and mms2 within the same pathway, but does not discem whether Srs2 functionç upstream of Rad6 and is required for al1 PRR, or if it is involved specifically within error-free PRR. If the latter is true, the srs2 mutation should have no effect on the increased mutagenesis observed in mms2 cells. As seen in Figure 19, while the mm2 single mutation causes an increase in the reversion frequency of a trp1-289 amber mutation up to 30 fold over wild type levels, the srs2 mutation suppresses the spontaneous mutator phenotype of mms2 cells. Thus we conclude that the mutator phenotype observed in mms2 cells is dependent on both REV3 and on SRS2.
The ability of the srs2 mutation to suppress the extreme UV sensitivities of rad6 and rad18 cells indicates that Srs2 is epistatically linked to Pm. If Srs2 is required to either prevent recombination-mediated replication restart, or to provide a function within RAD6-dependent PM, then it should be epistatically linked to other members of PRR. We have shown that srs2 is epistatic to mms2, which supports the placement of MMS2 within PRR. The srs2 and srs2 mms2 mutants display spontaneous mutation rates similar to wild type cells. This indicates that Srs2 may only be involved in error- free PRR. The reu3 srs2 double mutant was created to study the genetic interactions between srs2 and the error-prone PRR pathway. The srs2 mutant is more sensitive than the rev3 single mutant and the srs2 rev3 double mutant appeared to display a level of suwival similar to the srs2 single mutant (Fig. 20).
Therefore, srs2 is also epistatic to rev3. O 50 100 150 200 O 2 0 4 0 6 0 8 0 Dose (Jlm2) Minutes in 0.3% MMS
Figure 18: srs2 is epistatic to rnms2. The srs2, mms2, and srs2 mms2 mutants were exposed to vanous doses of UV (A) and MMS (B). The percent survival of the srs2 mms2 ( ) double mutant parallels the survival of the srs2 ( O ) single mutant, indicating that mms2( ) is hypostatic to srst. DBY747 ( a) wild type control. The results are an average of three independent experiments. Strain
Figure 19: The effect of the srs2 mutation on the spontaneous mutation rate of
mms2. Wild type cells, DBY747 (Strain l), display a low basal level of spontaneous mutagenesis compared to the mms2 mutant (Strain 2). The srs2 single mutant (Strain 3) displays a spontaneous mutation rate similar to wild type. The srs2 mutation in a mms2 background (Strain 4) reduces the spontaneous mutation rate to levels displayed by wild type and sr52 cells. The sensitivities of the mms2 and rm3 single mutants to MMS and UV were less severe compared to the phenotype of the srs2 mutant. The srs2 rad6 and srs2 rad18 mutants were as sensitive to MMS and W as the srs2 single mutant; SRS2 must control a critical step in PM, and when mutated establishes the phenotype of the known PRR mutants. To determine whether a srs2 mutation could suppress mms2 and reu3 mutants, the effeds of the srs2 mutation were tested on the mms2 rev3 double mutant. Gradient plates were used to study the effect of the srs2 mutation on the mms2 rm3 double mutant. As seen in
Figure 21, the srs2 mms2 rai3 triple mutant shares a similar level of suwival on
MMS as the srs2 single mutant, supressing the extreme sensitivity of the mms2 rev3 mutant. The resdts from the epistasis analysis indicate that mms2 and rev3 are hypostatic to srs2, and the double mutant is suppressed by the same mutation.
3.6.2. UV-induced Mutagenesis
The trpl-289 amber reversion assay detects spontaneous reversions at a relatively low basal level. The low frequency of spontaneous Trp+revertants in wild type DBY747 is suitable for studying alleles with a mutator phenotype, but does not allow for the assessrnent of alleles with anti-mutator phenotypes. rm3A cells have an anti-mutator phenotype in most reversion assays, due to the loss of mutagenic translesional bypass, but the trpI-289 reversion assay indicates that rev3A cells are similar to wild type. The trpl-289 reversion assay is able to differentiate between cells that are proficient and cells that are deficient in UV- induced mutagenesis since the basal level of reversion in DBY747 ceus is O 2 O 4 O 6 0 8 Minutes in 0.3% MMS
Figure 20: SRS2 and the mutagenic response. The srs2 mutation ( ) confers a
greater MMS sensitivity than the rev3 ( O ) single mutant. The % sumival of the srs2 reu3 ( ) double mutant was less than additive, cornpared to the
correspondhg single mutants, indicating that srs2 is epistatic to rev3. DBY747
( a)was wdas the wild type control. The resultç are an average of three independent experiments. 123456 Strain
Figure 21: Suppression of the mms2 rat3 MMS sensitivity by srs2. Compared to the DBY747 wild type strain (Strain l),mms2 (Strain 2). rev3 (Strain 3) and srs2
(Strain 4) show moderate sensitivities to 0.01% MMS. The mmd rev3 double mutant (Strain 5) is extrernely sensitive to MMS, but is suppressed by the additional çrs2 mutation (Strain 6). elevated with low doses of UV. 1was unable to address whether the srs2 mutation had a spontaneous mutator phenotype that was sirnilar to wild type cells, or if it conferred an anti-rnutator phenotype. Thus, 1wanted to use W- induced mutagenesis to differentiate between these two possibilities.
UV-induced mutagenesis was used to assess whether the srs2A mrns2A double mutant was defective in the mutagenic response or if it displayed wild type levels of trpl-289 reversion. As seen in Table 7, srs2A cells had similar levels of trpl-289 reversion at low doses of UV, similar to wild type cells. The redudion of spontaneous mutagenesis in the srs2A rnms2A double mutant cornpared to the mms2A single mutant, combined with the W-induced mutagenesis, indicates that Srs2 prevents the biased use of the mutagenic pathway when Mms2 is not present. Yowever, UV-induced mutagenesis is still dependent on a functional Rev3, since the srs2 rev3 UV-induced mutation rate is
ôbolished (Table 7).
3.6.3. Epistasis Between srs2 And The Other Members Of Error-Free PM
Al1 PRR mutants (error-free and error-prone) are hypostatically related to rad6A and radl8A. The fact that thesrs2 mutation suppresses rad6 and rad18 mutations indicaies it may fundion upsheam of both enzymes; however, its relationship with the other members of PRR have not been addressed. Epistasis analysis indicated that srs2 is epistatic to mms2 (Fig. 18) and to rm3 (Fig. 20).
The extreme MMS and UV sensitivity of the mms2 rev3 mutant is reduced by hirther mutating the srs2 gene; the triple mutant displays a level of W and
MMS induced killing sirnilar to the srs2 single mutant (Fig. 19). 1 wanted to test the effect of a srs2 mutation on the MMS sensitivity of the poZ30-46 and rad5 mutants.
Epistasis analysis with mutations of the proposed error-free PRR pathway indicates that the gene products of POL30, RAD5 and MMS2 do not cooperate within a single error-free pathway. The pol30-46 allele and the rad5 nul1 mutation confer a MMÇ sensitivity that is greater than the sensitivity of the srs2 single mutant, but the MM!3 sensitivity of pol30-46 and rad5 single mutants is suppressed by a mutation in srs2 (Fig. 22A and 22B). pol30-46 srs2 and rad5 srs2 mutants share the same level of MMS survival as the srs2 single mutant,
ùidicating that srs2 suppresses po130-46 and rad5 (Fig. 22A and 22B). The extreme MMS sensitivity of the poZ30-46 rad5 double mutant is suppressed by a third mutation in SRS2 with the triple mutant exhibiting a MMS sensitivity sirnilar to the srs2 single mutant (Fig. 22C).
3.6.4. SRS2 And Error-free Translesional Bypass
The RAD30 gene is part of the RAD6 PRR pathway, but the genetic data
(Fig. 15) and biochemical activity indicate that Rad30 may belong to its own error-free PRR pathway. Using the same rationale as in Section 3.8.3., 1 wanted to assess whether the independent activity of Rad30 was influenced by the activity of Srs2. A srs2 rad30 mutant was constructed to study the genetic relationship between the two genes.
Interestingly, the difference in UV sensitivity between rad30 mutant and wild type is more drastic than the sensitivity difference to MM, indicating a possible bias in lesion recognition and bypass (Fig. 23). Due to the weak MMS sensitivity of the rad30 mutant, it is not surprising that the rad30 srs2 is as sensitive to MMS as the srs2 single mutant. The sensitivity of the rad30 srs2 double mutant to W was as sensitive as the srs2 single mutant (Fig. 23). The epistatic relationship between rad30 and srs2 suggests that Rad30-rnediated translesiond bypass is also helicase-independent. The activity of Srs2 is obviously necessary for the activity of Rad6, Radl8, Mms2, Rev3, Rad5, Rad30 as well as Po130. Figure 22: The effect of the srs2 mutation on the members of error-free PRR.
The srs2A mutation was created in the pol30-46, rnd5 and po130-46 rad5 mutant backgrounds. The following strains were tested for their sensitivity to 0.1%
MMS: (A) Wild type PY39-0 ( a), PY39-46 (pol30-46 allele) ( O ), PY39srs2A ( .) and the double mutant PY39-46 srs2A ( ); (8) PY39 rad5A ( O ) and PY39 rad5A srs2A ( ); and (C) PY39-46 rad5A ( O ) and PY39-46 rad5A srs2A
( ). The srs2A mutation suppresses the MMS sensitivity caused by the po130-
46, rnd5A and po130-46 rad% mutations.
O 2 O 4 O 6 O a O O 50 100 150 200 Minutes in 0.3% MMS Dose (J/rn2)
Figure 23: Epistasis analysis between rad30 and srs2. The wild type ( a ), rad30
( ), srs2 ( O ) and rad30 srs2 ( ) mutants were treated with V~~OUSdoes of W (B) or with 0.3% MMS (A) for various lengths of time. The rad30 mutation does not display a significant sensitivity to MMS. The level of survival of the rad30 srs2 double mutant after treatment with UV is less than additive, indicating that srs2 is epistatic to rad30. 3.7. MMS2 Human Homologues
3.7.1. Identification of h2MMS2
The deduced S. cerevisiue Mms2 arnino acid sequence was used to search
the Human Genome Sciences Inc. (HGS) database for expressed cDNA sequence
tags (ESTs). With cutoffs that would elirninate all known Ubcs, two distinct
groups of cDNAs were identified. One group was soon found to be identical to
the CROC-2 isolated from the laboratory of a collaborator (Rothofsky & Lin,
1997),whereas a second group of cDNAs appeared to be novel, and were thus
named hMMS2. Arnong 19 initial putative hmS2clones, HGS36,109 contained
the longest cDNA (1.5 kb). Determination of the nucleotide sequence of its
entire cDNA insert revealed HGS36,109 to encode a 137 amino-aad polypeptide
that is highly homologous to Mms2. However, no translation start codon was
obsenred at the 5' end. The HGS database was subsequently searched for the
MI-length hMMS2 cDNA clone, which resulted in the isolation of pBShM2
(HGS1,273,684) from a human colon carcinoma cDNA library, containing 45
additional nucleotides with an in-frarne ATG, and surrounding nucleotide
sequences characteristic of a translation start codon (Fig. 24). It could encode a
145 amino aad, 16.4 kDa protein with 50.4% (69/137) sequence identity to
Mms2 throughout the entire length. In particular, the N-terminal halves of the two proteins are 65% (45/69) identical, encompassing two highly conserved
stretches with l3/ 13 and l3/ 16 arnino acid sequence identity (Fig. 25); the second stretch contains an identical potential phosphorylation site RlYSL for both cAMP-dependent and Ca+-calmodulin-dependentprotein kinases -GAAG A~CACAGGAMTAAAGTIT~AA~C~OTAGGC(3ACCGPACA MAVSTGVKVPRNFRLLEELEEGQKGVGDGT
GIT-mW'PGAAGATA'K VSWGLEDDEDMTLTRWTGMIIGPPRTNYEN
AGAATATATATAGC~TAGAAntTGGACCTAAATACCCAGAAGCTCCITCGIY3AC?TAGA?TIY;TAACAAAAA?TAATATGAACGGA RIYSKLVECGPKYPEAPPSVRFVTKINMNG
ATAAATAATK:CAC;'K;GGA~TGCCCCGAGCATACCA~TTAGCAAAATGGCAAAA'ITCATATAGCATT~ACmYlAA INNSSGMVDARSIPVLAKWQNSYSIKVVLQ
Figure 24: Nucleotide and deduced amino acid sequences of the hMMS2 cDNA (GenBank accession number AF049140). The 1,535 bp cDNA contains a single ORF encoding a putative 145 amino acid polypeptide. Two consensus polyadenylation signals (AATAAA) are underlined. Restriction sites EcoRI and SphI used in this study are indicated. (Kishimoto et al., 1985; Glass et al., 1986; Edelman et al., 1987). Like Mms2 and
CROC-1, the deduced hMMS2 protein does not contain a cysteine residue surrounded by the consensus sequences found in al1 Ubcs (Finley & Chau, 1991;
Jentsch, 1992). Recently, Dr. Xiao purified hMrns2 and has shown that it is not able to bind ubiquitin, ùidicating that hMms2 does not function as a Ubc. Our hMMS2 cDNA sequence (GenBank accession number AF049140) was found to be largely identical to D D Vit 1 (accession number X98091), isolated through differential display after Vitamin D, induction in human blood monocytes
(Fritsche et al., 1997). However, hMMS2 cDNA contains an additional 335 bp sequence in the 3' untranslated region compared to DD Vit 1. The same gene was also reported as a putative enterocyte differentiation promoting factor
EDPF-1 (accession number U62136). Northern hybridization with a hMS2 probe was performed by Barbara Chow,and the results indicate that this 1,535 bp hMMS2 cDNA likely represents the full length transcript.
3.7.2. Functional Cornplementation Of mms2 By hMMç2 And CROC-1
Given the strong homology between hMMS2, CROC-1 and Mm2, we asked if hMMS2 and CROC-1 are able to provide hiZMS2 functiow in yeast cells.
The yeast mms2 mutant is sensitive to killing by DNA damaging agents including MMS and UV. The IzMMS2 and CROC4B ORFs were cloned into a yeast expression vector under the control of an inducible GALl promoter. As seen in Figure 26, P,,,,-hMMS2 was able to rescue mms2 cells from killing by
MMS on a galactose (induction) plate, but not on the glucose (repression) plate, consistent with GALl promoter-controued gene expression. In contrast, P,,,,-
CROC-IB was unable to complement mms2 even on the galactose plate. CROC-10 IiMMS2 w CROC-ID hMMS2 Mm2
Figure 25: Sequence allignment of MmsZ and its human homologues.
Alignment of Mrns2, hMms2 and Croc-1. Shaded boxes indicate conserved amino acid residues. The failure of CROC-IB to complement mms2 could be due to either a la& of expression (rnRNA or protein instability), the amino acid sequence divergence between hMMS2 and CROC-1, or due to the additional N-terminal sequence unique to CROC-1B. To address these possibilities, we made a construd that produces a truncated CROC-1 comparable in length to hMMS2 and Mms2 under the control of a constitutive ADHl promoter (Fig. 27 & 28),
PADH1-CROC-1A80. As a reference, hMh/lS2 was also cloned into the same vector. Both PA,,,-WS2 and P,,,,CROC-lA80 were able to proted mms2 cells from killing by MMS and W (Fig. 27), implying that lack of complementation by CROC-18 is due to interference by its N-terminal sequence.
Although over-expression of the hurnan homologs Wly restored MMS resistance of the mms2 mutant to the wild type level, it did not provide additional resistance. This is consistent with the observation that over- expression of the MMS2 gene does not confer MMS resistance above the wild type level (Fig. B), suggesting that endogenous MMS2 is not a lirniting factor in the PRR pathway. Probably the most astonishing effect of the mms2 mutation is the remarkable increase in the spontaneous mutation rate, leading to a hypothesis that Mms2 functions in the error-free PRR that possibly suppresses the mutagenesis pathway. To see if the hurnan MM52 homologs are able to fully complement mms2 defects, hm2was cloned in a yeast expression vector suitable for the mutagenesis assay, and the spontaneous Trp' reversion rates of the wild type, mms2 mutant, and mms2 transformed with hMMS2 were determined. As seen in Table 8, the spontaneous Trp' reversion rate increased over 40-fold in the mms2 mutant; however, it was dramatically limited in the sarne cells expressing the hMMS2 gene. The 3.4-fold increase in the hMMS2 transfonnants compared to the wild type cells can be attnbuted to the loss of plasmid in a small population of the hMMS2 transformants (Table 8). YPD
YPD YPGal 0.03% MMS 0.03% MMS
Figure 26: Complemention of mms2 by hMMS2. The hMMS2 gene was doned next to the GALl promoter, allowing hMMS2 to be upregulated by plating the transformed cells on plates with galactose as the carbon source. The mms2 mutant strain is unable to grow on both YPD+MMS and YPGal+MMS plates.
The mmsî mutant transformed with pYES-hMMS2 is unable to grow on YPD plates in the presence of MMS, but can complernent the MMS sensitivity of mms2 on YPGal+MMS plates.
I 1 1 0.001 1 I O 2 0 4 0 6 0 8 O
Minutes in 0.3% MMS UV Dose (J/m 2)
Figure 27: Truncated CROC-1B is able to complement the mms2 mutant.
Transformants of the mms2 mutant harboring plasmids with either hMMS2 or the truncated CROC-IB (CROC-1,) were treated with MMS (A) and UV (B). The sumival was assessed to deterrnine the level of complementation for each construct. Cells were incubated in a selective medium pnor to killing experirnents to rnaintain the plasmids. ( a)DBY747 (wt), ( ) SBL (rnrnsZ), ( A ) SBL/pSCW-hM2, and ( O ) SBL/pSCW-ClA80. Results are an average of two independent experiments. Protein MMS
Figure 28: Complementation and homology overview of mms2 with hMMS2 and CROC-1 B. Filled boxes indicate two highly conserved motifs arnong al1 three proteins. Hatched boxes indicate regions that are moderately conserved arnong al1 three proteins. Dotted boxes indicate sequences that are only moderately conserved between CROC-1 and hMMS2. Open boxes represent sequences with no homology in other proteins. CHAPTER FOUR: DISCUSSION
4.1. Redefining The Rad6 Pathway
4.1.1. lMlMS2 1s Part Of The RAD6 PRR Pathway
Several lines of direct and indirect evidence support the assertion that the
MMç2 gene belongs to the RAD6 epistasis group. First, the mms2 nul1 mutant is
sensitive to killing by both MMS and UV at the level characteristic of a RAD6- pathway mutant. This is in contrast to the rad3-pathway mutants, which are
especially sensitive to UV and chernicals that produce structurally distorting lesions, but are marginally sensitive to MMS; and to the rad52-pathway mutants, which are extremely sensitive to ionizing radiations and MMS, but less sensitive to UV compared to mutants belonging to the other two pathways (Prakash,
1993; Friedberg et al., 1995; Xiao et al., 1996). Secondly, the amino acid sequence of Mms2 is homologous to al1 Ubcs and promotes ubiquitination when complexed with Ubcl3, an activity required for the RAD6 DNA repair pathway.
Finally, the mms2 mutation is hypostatic to both rad6 and rad1 8, but is additive to rad4 and rad50, suggesting that it is defective in the RAD6 pathway.
4.1.2. MMS2 And The Error-Free PRR Pathway
The dedaration that MMS2 functions in an error-free PRR pathway alternative to the error-prone pathway is based upon the following combined evidence: i) MMS2 belongs to the RAD6 epistasis group, ii) the mms2 mutant does not affect REV3-mediated spontaneous and UV-induced mutagenesis, and iii) the mms2 mutation is synergistic with the reu3 mutation. The placement of
MMS2 within the RAD6 PRR pathway is presented in Çection 4.1.1.
While the mutagenesis pathway is known to be mediated by REV3,
encoding the catalytic subunit of PolL as well as other genes such as REVl
(Larimer et al., 1989; Nelson et al., 1996b) and REV7 (Nelson et al., 1996a), little
is known about the error-free PRR pathway. Although the mms2 single mutant is only moderately sensitive to MMS and UV, our data indicate that the mms2
mutation is defective in the error-free PRR pathway. We ha~reshown that the
increased spontaneous mutation rate observed in the mms2 mutant is completely
dependent on the hctional RE V3, indicating that when error-free PRR is
dysfunctional, al1 the spontaneous lesions could be channelled to the mutagenesis pathway. To further support Our argument, the mms2 rm3 double
mutant is extremely sensitive to killiiig by W. Simultaneous defects in both
error-free PRR (rnms2) and mutagenesis (rev3)branches display MMS and W
sensitivities that approach the level obsewed in rad18 mutants, known to be
defective in both pathways. The definition of the error-heee PRR pathway was further secured by the
phenotype of the rnd6,,, mutation combined with our mms2 results; Watkins et al. (1993) demonstrated that the N-terminus of Rad6 is required for error-free
PRR but not foi-mutagenesis. mms2 is hypostatic to this N-terminal deletion (rad6,,,), providing another putative mernber of the newly, ill-defined PRR subpathway. 4.2. Organization Of Error-Free PRR. Epistasis analysis was used to establish the genetic relationship of the genes assigned to the RAD6 error-free pathway. The pol30-46 allele of the yeast
PCNA gene POL30 is sensitive to UV and MMS, displays an increase in spontaneous and UV-induced mutagenesis, and has been shown to be defective in PRR (Ayyagari et al., 1995; Torres-Ramos et al., 1996). Surprisingly, the mms2 pol30-46 double mutant is more sensitive to UV and MMS than either single mutant; these results suggest that there is more than one pathway defective in the repair of MMS and W in the double mutant. Another proposed member of the RAD6-mediated error-free PRR is RADS; the rad5 mutant is hypostatic to rad6 and radl8, synergistic with rai3 and displays a mutator phenotype (Johnson et al., 1992). These results fit the aiteria of a gene involved in error-free PRR.
However, the expected epistasis between RAD5 and MMS2 was not obsewed; instead, the rad5 mms2 double mutant has reduced survival after UV or MMS exposure compared to mms2 or rad5. Therefore, the same conclusion can be drawn with rad5 and mms2, as was concluded with mm2and pol30-46: RAD5 and MMS2 are part of two different sub-pathways.
The Rad30 polymerase does not cooperate with Mms2 in error-free repair. Firstly, epistasis analysis indicates that there is a more severe killing effect in the double mutant than the single mutants, providing evidence of involvement in two different error-free PRR pathways (Fig. 15). Secondly, prelirninary work by other groups and by our lab suggests that Rad30 is specific for tolerating damage created by UV (Fig. 15), since the rad30 mutant exhibits wild type levels of MMS sensitivity. Such a specific activify may be mediated by Rad6 and Radl8, but appears to be sepxated from the other Rad6 sub- pathways.
Further work was completed in our lab to address genetic relationships between po130-46 and rad5, poZ30-46 and rad30, as well as rad5 and rad30. The data obtained by Barbara Chow and Michelle Hanna for the above mentioned epistasis analysis indicate that each gene belongs to a separate pathway. The pol30-46 rad5 (Fig. 22C) exhibits additive killing compared to the single mutants
(Fig. 22A and 228); the sarne additive response is observed in the exposure of pol30-46 rad30 and rad5 rad30 double mutants to UV and MMS. This information is perplexing; the interpretation that multiple error-free sub-pathways exist under the control of Rad6 does not seem logical. The role of error-free PRR is to resolve stalled replication forks, allowing the resumption of replication without removal of the damage, without the integration of a mutation in the newly synthesized DNA. It does not seem likely that the activity of a single enzyme downstream from the Rad6-Rad18 heterodimer can provide al1 the activities required for replication restart and gap-filling. Assessing the capabilities of each enzyme associated with error-free PRR has not provided any clues as to how the pathway fulfills the gap-filling activity. A possible explmation for the additive sensitivities of the mutants could be due to multiple functions of each enzyme in other branches of the PRR pathway. Our ignorance about the structure of darnaged DNA may hinder our interpretations of epistasis analysis. Although
PRR is involved in processing DNA damage encountered by a replication fork, there rnay be a universal activity required for al1 types of DNA damage repair.
If any of the error-free PRR genes had functions within error-free and error- prone PRR, it may explain why the epistasis analysis of the error-free PRR mutants was additive.
4.3. MMS2 In Spodation And Protein Degradation
There appears to be a small discrepancy about the level of involvement of
Rad6 in sporulation; depending on the strain background, rad6 mutants display sporulation efficiency that is 0.040 % of wild type diploids (Momson et al.,
1988). Despite the abiIity of certain rad6/rad6 diploids to achieve 4.0 % spomlation efficiency, these levels are very low and indicate that Rad6 is an important factor in spomlation. mms2/mms2 diploids have a moderate defect in sporulation. The mms2 diploids 1made were derived kom FY86 x XS-803-3A backgrounds, but H. Ulrich has shown similar levels of spomlation efficiency between mms2 and wild type in YWOl and YM02 backgrounds (personal communication; Max-Planck-Institute for Biochernistry). Assessing the sponilation levels in rad6 and mms2 diploids does not darify how these two enzymes relate in different cellular processes. The necessity for Rad6 in sporulation and the partial involvement of Mms2 could be explained by two different hypotheses. First, Mmç2 may be employed as a novel E3 enzyme to help Rad6 define its spomlation target, histone H2B (Robzyk et al., 2000).
Without Mms2, the efficiency of Rad6 is much reduced, resulting in a moderate spomlation defect. Another possible explanation for the sporulation defect codd be that Mrns2 provides an alternative Ubc function. Lf this alternative Ubc activity exiçts, then the moderate sponilation defed observed in mms2 cells indicates that the Rad6-dependent Ubc activity is preferred. Mms2 appears to have an indirect role in targetting proteins for
degradation. The difference between the stability of Leu-pgal and Arg-pgal in
wild type and rad6 cells provides convincing evidence that Rad6 is very
important in the ubiquitin N-degron pathway. The Leu-pgal and Arg-pgal
constructs have slightly increased half lives in mms2 mutants, but do not exhibit
the stability obsewed in rad6 cells. Hence, the cellular effect of the mms2
mutation appears to be similar for both sporulation and protein ubiquitination.
As mentioned for spodation, the involvement of Mms2 in protein degradation
can be expiained in a Rad6-dependent or Rad6-independent marner.
In determinhg the involvement of Mms2 in telomere silencing, the
results are less arnbiguous as they were in the spodation and protein
degradation studies. Phenotypic evidence indicates that Rad6 is one of the
many enzymes required for telomere position effect, while our data suggests
that Mms2 is not involved.
The involvement of Mms2 in the adivities mediated by Rad6 is perplexing. How could Mms2 play a role in error-free PRR,have an effect on sporulation, a minor effect on protein degradation, without an involvement in telomeric silencing? Mms2 may fulfill a role as an accessory protein to Rad6, with an activity that can be replaced by another protein. It may also be possible that in the absence of MMS2, normal cellular activities are dismpted; without
Mms2 (and error-free PRR), subsequent disruptions to the genome may affect other pathways in the cell. Furthemore, Mms2 may provide an alternative Ubc pathway, independent of Rad6. Until the biochemistry of Mms2 is understood, the function of Mms2 and how it relates to Rad6 can only be postulated. 4.4. MMS2 And LIBC13
Ub diain formation is based on Ub moities joined together through
LysineClycine Iinkages. Ub chains with the Lys48 (~48)Gly76 (G76)linkage are recognized by the 26s proteosorne, and is tkrefore of great biological significance for protein degradation (Chau et al., 1989). Ub contains another five lysine residues, and four of these can form Ub chains. Spence et al (1995) studied Ub mutants and were able to distinguish the biological differences between the vanous linkages. As mentioned in Section 1.6.2.2., the Ub~63 linkage is involved in DNA darnage tolerance.
Hofmann and Pickart (1999) isolated Ubcl3, a novel Ubc enzyme, in a screen for the Ubc responsible for UbK63 linkage formation. They further discovered that Ubcl3 required the presence of another protein for its Ubc activity; the second protein is Mms2 (Hofmann & Pickart, 1999). Ubcl3 and
Mrns2 form a heterodimer that facilitates Ub~63linkages. Epistasis analysis from our lab have shown that ubcl3 is epistatic to mm2 and belongs to the error-free PRR pathway (Brusky et al., 2000), confirming and extending the findings of Hofrnann and Pickart (1999) and of Spence et al (1995) that Ub~63has a role in the RAD6 pathway.
The addition of Ubcl3 to the RAD6 pathway helped define the role of
Mrns2 wîthin PRR, but increased the complexity with having two Ubc enzymes within one pathway. Of great importance in delineating the Rad6 pathway is the identification of the targets of Rad6/Rad18 mediated ubiquitination and of
Ubc13/Mms2 mediated ubiquitination. 1believe the identification of the ubiquitinated target of Ubcl3/Mms2 can be detennined by establishing the cellular protein(s) bound to Ub via the unique Ub~63linked polyubiquitin chah It is not known whether Ubcl3 is involved in spodation and protein degradation, possibly providing the alternative Ubc activity in cooperation with Mms2.
Recent work by Ulrich & Jentsch (2000) uncovered novel interactions within error-free PRR. Using two-hybrid and immunoprecipitation assays
Uirich and Jentsch (2000) observed the homodimerization of both Rad18 and
Rad5 and the heterodimerization of Rad18 and Rad5. Rad5 is also able to interact with the Ubcl3:Mms2 dimer, possibly coordinating the interaction between Ubcl3:Mms2 and Radl8:Radb. The Rad5:Ubcl3 interaction is dependent on the Rad5 RING finger motif, a novel type of zinc finger (Freemont et al., 1991). The RING finger motif of important cellular proteins (BRCAl, Siah- 1, Prajal, TRC8, NF-XI, kf-1 and cal)has been implicated in mediating E2- dependent ubiquitination (Lorick et al., 1999). Further analysis of Rad5 may indicate that the RING finger is involved in ubiquitination as well as in protein:protein interactions.
4.5. The Role Of Srs2 In DNA Damage Tolerance
It is assumed that Rad6 is the enzyme that initiates and mediates the major DNA damage tolerance pathway in S. cermisine. The necessity for Rad6 in
PRR is not disputed, but is it the enzyme that provides the primary hction within PRR? Isolation of SRS2 has raised many questions as to the fundional order of the defined PRR enzymes. The genes that comprise PRR were placed within this pathway by epistasis analysis with rad6 and radl8. In DNA repair studies, epistasis analysis is an important genetic tool for placing newly identified genes into known repair pathways based on the phenotype of the double mutant; however, epistasis analysis is limited in that it camot determine the order of each specific enzymatic abivity within a pathway. In this study the
UV and MM!3 sensitivities of po13046, rad5, rev3, and mms2 are determined by the presence or absence of Srs2. In interpreting these data, two questions must be addressed. First, what function does Srs2 perform that accounts for these epistasis results? Second, if Srs2 is required for PRR, why is it moderately sensitive to DDNA darnage, and not extremely sensitive as is seen with rad6 and rad18 cells.
The srs2 mutation suppresses the DNA darnage sensitivity of PRR mutants, and has been interpreted as acting as a pathway switch between PRR and recombination repair. No further interpretation has been offered as to how the helicase could promote RAD6-mediated PRR. Within the context of a stalled replication fork, how does the ce11 process this structure to prevent a lethal event? In the absence of Rad30 and Rev3 there still exists an unknown error-free
PRR activity that is independent of recombination repair. The Srs2 helicase must act at the stalled replication fork, possibly as a helicase, to create a template for the enyzmes of PRR to function. Loss of Srs2 does not affect the spontaneous or W-induced mutation rates, which contradicts the idea of it acting upsheam of the mutagenic polymerase. The Rad6:RadlB heterodimer must provide a critical fmction downstream of Srs2. Without either gene, cells become defective in both error- free and error-prone PRR and are incredibly sensitive to DNA damage.
Whatever role Srs2 plays in PRR, its resolution requires the adivity of the Rad6:RadlS heterodimer. In the presence of Srs2, Rad6 and Radl8, what controls the next step? What promotes error-free PRR and prevents mutagenesis? Currently, the genetic data indicates that there are many genes involved downstream of Srs2, Rad6 and Radl8, but their mode of action is unknown. Until more information is gathered on the biochemistry of error-free
PRR enzymes, any hypotheses on the adivity of error-free PRR is difficult to analy ze.
4.6. Evolutionary Conservation
I was involved in the early work with hMMS2 and CROCI, which is now carried out by Carolyn Ashley, Parker Anderson and our collaborators. The identification of human sequence homologues of MMS2 was exciting, but it was not until hctional complementation was observed that we really envisioned the potential of the project. Functional complementation of the yeast mms2 mutant by hMMS2 suggests that hMMS2 may have a role in human PRR.
However, CROC1 does not complement the MMS sensitivity of the yeast mms2, and sequence analysis cannot resolve whether there is functional conservation between yMMS2 and CROCI. CROCl contains a unique N-terminal sequence, and if removed the truncated CROCl (CROCI,,,) can complement the MMS sensitivity of the yeast rnrns2A sixain. The sequence similarity between the two proteins could indicate that they have a sirnilar biochernical activity, but their cellular roles may be quite different. The N-terminus of CROC2 rnay be responsible for defing the cellular activity of CROC1, which may be unique from hMMS2.
In collaborative efforts we have shown that CROC2 appears to be upregulated in immortal cells (Xiao et al., 1998; Ma et dl., 1998). Northem analysis of WS2and CROCl transcripts show that their expression is elevated in certain cancer ceU lines (Xiao et al., 1998). More work must be conducted to study the role of these genes in human cells, and whether either has a roie in human PRR.
It is important to study the evolution of PRR. The prernise of the entire ' pathway is to prevent ce11 death, at any cost. How does such a pathway function in an organism where mutation may not be advantageous? With a strict surveillance system and with the option to apoptose, where does PRR fit in for a multi-cellular organism? Has it evolved to serve a different purpose in the cell? The continued research on hMMS2 and CROCl may shed some light on whether either gene is involved in PRR or if they hilfill another cellular role.
Despite the rnystery surrounding the function of both CROCl and hMMS2, more information on these genes may be of use in the field of health care in the future. Re£erences:
Aboussekhra, A., Chanet, R., Zgaga, Z., CassierChauvat, C., Heude, M., & Fabre, F. (1989) RADH, a gene of Saccharomyces cwevisiae encoding a putative DNA helicase involved in DNA repair. Characteristics of radH mutants and sequence of the gene. Nucleic Acids Res. 17: 7211-7219.
Aboussekhra, A., Chanet, R., Adjiri, A. & Fabre, F. (1992) Semidominant Suppressors of Srs2 Helicase Mutations of Saccharomyces cereuisiae Map in the RAD51 Gene, Whose Sequence Predicts a Protein with Similarities to Prokaryotic RecA Proteins. Mol. Cell. Biol. 12: 3224-3234.
Ahne, F. Jha, B. & Edcardt-Schupp, F. (1997) The RAD5 gene produd is involved in the avoidance of non-homologous end-joining of DNA double strand breaks in the yeast Saccharomyces cerevisiae. Nzicleic Acids Res. 25: 743-749.
Altschul, Ç.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic Local Alignment Search Tool. J. Mol. Biol. 215: 403-410.
Aparicio, O.M., Billington, B.L., & Gottschling, D.E. (1991) Modifiers of position effed are shared between telomeric and silent mating-type loci in S. cerevisine. Ce21 66: 1279-1287.
Araujo, S.J. & Wood, RD. (1999) Protein complexes in nucleotide excision repair. Mut. lies. 435: 23-33.
Armstrong, J.D., Chadee, D.N. & Kunz, B.A. (1994) Roles for the yeast RAD18 and~4~52DNA repair genes in UV mutagenesis. Mut. es. 315: 281- 902
Ayyagan, R., Impellizzeri, K.J., Yoder, B.L., Gary, S.L. & Burgers, P.M.J. (1995) A Mutational Analysis of the Yeast Proliferating Ce11 Nuclear Antigen Indicates Distinct Roles in DNA Replication and DNA Repair. Mol. Cell. Biol. 15: 4420-4429.
Bacher Reuven, N., Arad, G., Maor-Shoshani, A. & Livneh, 2. (1999) The Mutagenesis Protein UmuC is a DNA Polymerase Activated by UmuD', RecA, and SSB and 1s Specialized for Translesional Replication. 1. Biol. Chem. 274: 31763-31766.
Bahair, A., Finley, D. & Varshavsky, A. (1986) In vivo half-life of a protein is a function of its arnino-terminal residue. Science 234: 179-186.
Bailly, V., Lamb, J., Sung, P., Prakash, S. & Prakash, L. (1994) Specific complex formation between yeast RAD6 and RAD18 proteins: a potential mechanism for targeting RAD6 ubiquitin-conjugating adivity to DNA damage sites. Gmes Dm. 8: 811-820. Bailly, V., Prakash, S. & Prakash, L. (1997a) Domains Required for Dimerization of Yeast Rad6 Ubiquitin-Conjugating Enzyme and Rad18 DNA Binding Protein. Mol. Cell. Biol. 17: 4536-4543.
Bailly, V.,Lauder, S., Prakash, S. & Prakash, L. (1997b) Yeast DNA Repair Proteins Rad6 and RadlS Form a Heterodimer That has Ubiquitin Conjuga ting, DNA Binding, and ATP H ydrol ytic Activi ties. /. Biol. Clzeni. 272: 23360-23365.
Baynton, K., Bresson-Roy, A. & Fuchs, R.P.P. (1998) Analysis of Damage Tolerance Pathways in Snccharornyces cermisine: a Requirement for Rev3 DNA Polymerase in Translesional Synthesis. Mol. Cd.Biol. 18: 960-966.
Berben, G., Dumont, J., Gilliquet, V., Bolle, P.A., & Hilger, F. (1991) The YDp plasmids: a uniform set of vectors bearing versatile gene disruption cassettes for Snccl~nroniycescerevisinr. Yenst 7: 475-477.
Berends, W. (1961) On the dimerisation in nucleic acids as a result of ultraviolet irradiation. 1. Clzenz. Phys. 58: 1034-1040.
Boeke, J.D., Trueheart, J., Natsoulis, G. & Fink, G.R. (1987) 5-Fluoro-orotic acid as a selective agent in yeast rnolecular genetics. Mrtkods Enzymol. 154: 164-75
Burckhardt, S.E., Woodgate, R., Scheuermann, R.H. & Echols, H. (1988) UmuD mutagenesis protein of Escherichin coli: overproduction, purification, and cleavage by RecA. Proc. Nntl. Acnd. Sci. USA 85: 1811-1815.
Cassier-Chauvat, C. & Fabre, F. (1991) A similar defect in UV-induced mu tagenesis conferred by the rnd6 and rad 18 mu ta tions of Sncclinroniycrs cerevisine. Mrr f. Res. 254: 247-253.
Chan, G.L., Doetsch, P.W., & Haseltine, W.A. (1985) Cyclobutane pyrimidine dimers and (6-4) photoproducts block polymerization by DNA polvmerase 1. Bioclzernistn~ 24: 5723-5728.
Chanet, R., Heude, M., Adjiri, A., Maloisel, L. & Fabre, F. (1996) Semidominant Mutations in the Yeast Rad51 Protein and Their Relationships with the Srs2 Helicase. Mol. Cell. Biol. 16: 4782-4789.
Chau,V., Tobias, J.W., Bachrnair, A., Marriott, D., Ecker, D.J., Gonda, D.K., & Varshavsky A. (1989) A multiubiquitin chain is confined to a specific lysine in a targeted short-lived protein. Science 243: 1576-1583.
Chaudhury, A.M. & Smith, G.R. (1985) Role of Escherichin coli RecBC enzyme in SOS induction. Mol. Gen. Genet. 201: 525-528.
Chung, C.T., Niemela, S.L., & Miller, R.H. (1989) One-step preparation of competent Escherichin coli: transformation and storage of bacterial cells in the same soIution. Proc. Nntl. Acod. Sci. U S A 86: 2172-2175. Cleaver, J.E. (1968) Defective repair replication of DNA in xeroderma pigmentosum. Nature 218: 652656- Cleaver, J.E. (1972) Xeroderma pigrnentosum: variants with normal DNA repair and normal sensitivity to ultraviolet light. J. Invest. Dermtol. 58: 124128.
Courcelle, J., CarswellCrumpton, C., & Hanawalt, P.C. (1997) recF and recR are required for the resumption of repLication at DNA replication forks in Escherichia coli. Proc. Natl. Acad. Sci. USA 94: 37143719.
Courcelle, J. & Hanawalt, P.C. (1999) Re& and RecJ process blocked replication forks pnor to the resumption of replication in UV-irradiated Escherichia coli. Mol. Gen. Genet. 262: 543-551.
Cox, B.S. & Parry, J.M. (1968) The isolation, genetics and sumival characteristics of ultraviolet light-sensitive mutants in yeast. Mu tat. Res. 6: 37-55.
Cox, B.S. & Game, J-C. (1974) Repair systems in Saccharomyces. Mutat. Res. 26: 257-264.
Cox, M.M. (1997) Recombinational crossroads: Eukaryotic enzymes and the lirnits of bacterial precedents. Proc. Natl. Acad. Sci. USA 94: 1176411766.
Croteau, D.L. & Bohr, V.A. (1997) Repair of Oxidative Damage to Nuclear and Mitochondrial DNA in Mammalian Cells. J. Biol. Chem. 272: 25409-25412.
de Laat, W.L., Jaspers, N.G.J. & Hoeijmakers, J.H.J. (1999) Molecular mechanism of nucleotide excision repair. Genes Dai. 13: 768-785.
DiCaprio, L. & Cox, B.S. (1981) DNA synthesis in UV-irradiated yeast. Mut. Res. 82: 69-85.
Dohrnen, R.J., Madura, K., Bartel, B. & Varshavsky, A. (1991) The N-end rule is mediated by the UBC2 (RAD6)ubiquitin-conjugating enzyme. Proc. Natl. Acad. Sci. USA 88: 7351-7355.
Dubiel, W. & Gordon, C. (1999) Ubiquitin pathway: Another link in the polyubiquitin chah? Ctm. Biol. 9: R554-R557
Edelman, A.M., Blumenthal, D.K. and Krebs, E-(1987) Protein serine/threonine kinases. Annu. Rai. Biochem. 56: 567-613.
Ellison, K.S., Gwozd, T., Prendergast, J.A., Paterson, M.C. & Ellison, M.J. (1991) A Site-directed Approach for Consmicting Temperaturesensitive Wbiquitin-conjugating Enzymes Reveals a Ce11 Cycle Function and Growth Function for RAD6. J. Biol. Chem. 266: 24116-24120.
Finley, D. and Chau, V. (1991) Ubiquitination. Annn. Rm. Cell. Biol. 7: 25-69. Flores-Rozas, H. & Kolodner, R.D. (2000) Links between replication, recombination and genome instability in eukaryotes. Trends Bioch. Sci. ?: 196-200.
Friedberg, E.C., Walker, G. & Siede, W. (1995) DNA Repair and Mutagenesis (Am. Soc. Microbiol., Washington, DC).
Fritsche, J., Rehli, M., Krauçe, S.W., Andreesen, R. and Kreutz, M. (1997) Molecular cloning of a ldpha,E-dihydroxyvitamin D3-inducible transcript (DDVit 1) in human blood monocytes. Biochem. Biophys. Res. Contmun. 235: 407-412.
Ganesan, A.K. (1974) Persistence of pyrimidine dimers dunng postreplication repair in ultraviolet light-irradiated Escherichia coli. J. Mol. Biol. 87: 103- 119.
Gerik, K.J., Li, X., Pautz, A. & Burgers, P.M.J. (1998) Charactenzation of the Two Small Subunits of Sacchnrornyces cereuisiae DNA Polymerase 6. J. Biol. Chem. 273: 19747-19755.
Gibbs, P.E.M., Kilbey, B.J., Bane rjee, S.K, & Lawrence, C. W. (1993) The Frequency and Accuracy of Replication Past a Thymine-Thymine Cydobutane Dirner are very Different in Saccharomyces cereuisiae and Escherichia coli. J. Bacferiol. 175: 2607-2612.
Gietz, R.D. & Sugino, A. (1988) New yeast-Escherichia coli shuttle vectors conçtructed with in vitro mutagenized yeast genes ladcing six-base pair restriction sites. Gene 74: 527-534.
Giot, L., Chanet, R., Simon, M., Facca, C., & Faye, G. (1997) Involvement of the Yeast DNA Polyrnerase 6 in DNA Repair in Vivo. Genetics 146: 1239- 1251. Glass, D.B., El-Maghrabi, R. and Pilkis, S.J. (1986) Synthetic peptides conesponding to the site phosphorylated in 6-phosphofmcto-2- kinase/ midose-Z,&bisphosphatase as substrates of cyclic nucleotide- dependent protein kinases. J. Biol. Chem. 261: 2987-2993.
Gottschling, D.E., Aparicio, O.M., Billington, B.L. & Zakian, V.A. (1990) Position Effed at S. cerevisiae Telomeres: Reversible Repression of Pol II Transcription. Cell 63: 751-762.
Hartwell, L.H. & Weinert, T.A. (1989) Checkpoints: controls that ensure the order of ce11 cycle events. Science 246: 629-634.
Hersko, A. & Ciechanover, A. (1998) The ubiquitin system. Annu. Rev. Biochem. 67: 425-479. Higginç, NP., Kato, K. & Strauss, B. (1976) A Mode1 for Replication Repair in Mamrnalian Cells. J. Mol. BioZ. 101: 417-425.
W, J., Donalk, KA., Griffith, D.E. & Donald, G. (1991) DM=-enhanced whole ce11 yeast transformation. Nud. Acids Res. 19: 5791.
Hoffman, CS. & Winçton, F. (1987) A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli. Gene 52267-272.
Hofmann, R.M. & Pickart, C.M. (1999) Noncanonical MMSZEncoded Ubiquitin- Conjugating Enzyme Functions in Assembly of Novel Polyubiquitin Chahs for DNA Repair. CelL 96: 645-653.
Holliday, R. (1964) A mechanism for gene conversion in fungi. Genet. Res. 5: 282-304.
Huang, H., Kahana, A., Gottschling, D.E., Prakash, L. & Liebman, S.W. (1997) The Ubiquitin-Conjugating Enzyme Rad6 (Ubc2) is Required for Silencing in Saccharomyces cerpuisiae. Mol. Cell. Biol. 17: 66936699.
Jaruga, P., & Dizdaroglu, M. (1996) Repair of products of oxidative DNA base darnage in human cells. Nucleic Acids Res. 24: 1389-1394.
Jentsch, S., McGrath, J.P. & Varshavsky, A. (1987) The yeast DNA repair gene RAD6 encodes a ubiquitin-conjugating enzyme. Nature 329: 131-134.
Jentsch, S. (1992) The ubiquitin-conjugation system. Annu.. Rev. Genet. 26: 179- 207.
Johnson, R.E., Henderson, S.T., Petes, T.D., Prakash, S., Bankmann, M. & Prakash, L. (1992) Saccharomyces cerevisiae RADS-Encoded DNA Repair Protein Contains DNA Helicase and Zinc-Binding Sequence Motifs and Affects the Stability of Simple Repetitive Sequences in the Genome. Mol. Cell. Biol. 12: 3807-3818.
Johnson, RE., Prakash, S. & Prakash, L. (1994) Yeast DNA Repair Protein Rad5 That Promotes Instability of Simple Repetitive Sequences is a DNA- dependent ATPase. J. Biol. Chern. 45: 28259-28262.
Johnson, R.E., Kondratick, CM., Prakash, S. & Prakash, L. (1999a) hRAD3O Mutations in the Variant Form of Xeroderma pigmentosum. Science 285: 263-265.
Johnson, R.E., Washington, M.T., Prakash, S. & Prakash, L. (1999b) Bridging the gap: A family of novel DNA poiymerases that replicate faulty DNA. Proc. Natl. Acad. Sci. USA 96: 12234-12226.
Jonson, 2.0.& Hubscher, U. Proliferating ce11 nuclear antigen: more than a clamp for DNA polymerases. Bioessays 19: 967-975. Kadyk, L.C. & Hartwell, L.H. (1993) Replication-Dependent Sister Chromatid Recombination in rad1 Mutants of Sacchromyces cereuisiae. Genetics 133: 469-487.
Kishimoto, A., Nishiyarna, K., Nakanishi, H., Uratsuji, Y., Nomura, H., Takeyama, Y. and Nishizuka, Y. (1985) Studies on the phosphorylation of myelin basic protein by protein kinase C and adenosine 3':5'- monophosphate-dependent protein kinase. J. Biol. Chem. 260: 12492- 12499.
Koegl, M., Hoppe, T., Schlenker, S., Umch, H.D., Mayer, T.U., & Jentsch, S. (1999) A novel ubiquitination factor, E4, is involved in multiubiquitin diain assembly. Cell 96: 635-644.
Kogoma, T. (1997) 1s RecF a DNA Replication protein? Proc. Natl. Acad. Sci. USA 94: 3483-3484.
Lanzov, V., Stepanova, L, & Vùiogradskaja, G. (1991) Genetic control of recombination exchange frequency in Escherichia coli K-12. Biochimie. 73: 305-312.
Larimer, F.W., Perry, J.R. & Hardigree, A.A. (1989) The REVl gene of Saccharomyces cereuisiae: isolation, sequence, and fundional analysis. 1 Bncteriol. ln: 230-237.
Larson, K.L. & Strauss, B.S. (1987) Infiuence of template shandedness on in vitro replication of mutagen-damaged DNA. Biochemistry 26: 2471-2479.
Lawley, P.D. & Phillips, D.H. (1996) DNA adducts from chemotherapeutic agents. Mutation Research 355: 13-40.
Lawrence, C.W. & Christensen, R. (1976) W mutagenesis in radiation-sensitive strains of yeast. Genetics 82: 207-232.
Lawrence, C.W. & Christensen, R.B. (1979) Metabolic suppressors of trimethoprim and ultraviolet light sensitivities of Saccharomyces cereuisiae rad6 mutants. J. Bacteriol. 139: 866-887.
Lee, G.S.F., Blonsky, K.S., VanOn, D.L., Savage, E.A., Morgan, A.R., and VonBorstel, R.C. (1992) Base Alterations in Yeast hduced by Alkylating Agents with Differing Swain-Scott Substrate Constants. J. Mol. Biol. 223: 617-626.
Lehmann, A.R., Kirk-Bell, S., Arlett, C.F., Paterson, MC., Lohrnan, P.H.M., de Weerd-Kastelein, E.A. & Bootsma, D. (1975) Xeroderma pigmentosum cells with normal levels of excision repair have a defect in DNA synthesis after UV-irradiation. Proc, Natl. Acad. Sci. USA 72: 219-2î3. Lemontt, J.F. (1971) Mutants of Yeast Defective in Mutation Induced by Ultraviolet Light. Genetics 68: 21-33.
Lengauer, C., Kinzler, K. W. & Vogelstein, B. (1998) Genetic instabilities in human cancers. Nature 396: 643-649.
Liefshitz, B., Steinlauf, R., Friedl, A., Eckardt-Schupp, F. & Kupiec, M. (1998) Genetic interactions between mutants of the 'error-prone' repair group of Saccharomyces cerevisiae and their effect on recombination and mutagenesis. Mut. Res. 407: 135-145.
Lindahi, T. (1993) lnstability and decay of the primary structure of DNA. Nature 362: 709-715.
Ma, L., Broomfield, S., Lavery, C., Lin, S.L., Xiao, W. & Bacchetti S. (1998) Up- regulation of CIR1 /CROC1 expression upon cell imrnortalization and in turnor-derived human ceIl lines. Oncogene 17: 1321-1326.
Maniatis, T., Fritsch, E.F., & Sambrook, J. (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Maor-Shoshani, A., Bacher Reuven, N., Tomer, G. & Livneh, Z. (2000) Highly mutagenic replication by DNA polymerase V (UmuC) provides a mechanistic basis for SOS untargeted mutagenesis. Proc. Natl. Acad. Sci. USA 97: 565-570.
Masutani, C., Araki, M., Yarnada, A., Kusumoto, R., Nogimori, T., Maekawa, T., Iwai, S. & Hanaoka, F. (1999a) Xeroderma pigmentosurn variant (XF-V) correcting protein from HeLa cells has a thymine dimer by-pass DNA polymerase activity. EMBO 18: 3491-3501.
Masutani, C., Kusumoto, R., Yamada, A., Dohmae, N., Yokoi, M., Yuasa, M., Araki, M., Iwai, S., Takio, K. & Hanaoka, F. (1999b) The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase q. Nature 399: 700-704.
McDonald, J.P., Levine, A.S. & Woodgate, R. (1997) The Saccharomyces cerevisiae RAD30 Gene, a Homologue of Escherichin coli dinB and umuC, is DNA Damage Inducible and Functions in a Novel Error-Free Postreplication Repair Mechanism. Genetics 147: 1557-1568.
Michel, B., Ehrlich, S.D. & Uzest, M. (1997) DNA double-strand breaks caused by replication arrest. EMBO 16: 430-438. ModCch, P. (1994) Mismatch Repair, Genetic Stability, and Cancer. Science 266: 1959-1960. Morrison, A., Miller, E.J. & Prakash, L. (1988) Domain Structure and Functional Analysis of the Carboxyi-Terminal Polyaadic Sequence of the RAD6 Protein of Saccharomyces cerevisiae. Mol. Cell. Biol. 8: 1179-1185.
Nasmyth, K. (1996) Putting the Ce11 Cycle in Order. Science 274: 1643-1645.
Nelson, J.R., Lawrence, C.W. & Hinkle, D.C. (1996a) Thymine-Thymine Dimer By-passby Yeast DNA Polymerase <. Science 272: 1646-1649.
Nelson, J.R., Lawrence, C.W. & Hinkle, D.C. (1996b) Deoxycytidyl tram ferase activity of yeast REVl protein. Nature 382: 729-731.
Newlon, CS. (1996) DNA Replication in Yeast, pp873-914, in DNA Replication in Eukaryotic Cells, M.L. DePamphilis, ed. (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY).
Obe, G., Johannes, C., & Schulte-Frohlinde, D. (1992) DNA doublestrand breaks induced by sparsely ionizing radiation and endonudeases as critical lesions for ce11 death, chromosomal aberrations, mutations and oncogenic transformation. Mu tngoiesis 7: 3-12.
Paulovich, A.G., Margulies, RU.,Garvik, B.M. & Hartwell, L.H. (1997a) RAD9, RAD17,and RAD24 are Required for S Phase Regdation in Saccharomyces cerevisiae in Response to DNA Damage. Gmetics 145: 45-62.
Paulovich, A.G., Toczyski, D.P. & Hartwell, L.H. (199%) When Checkpoints Fail. Ce11 88: 315-321.
Paulovich, A.G., Amour, C.D. & Hartwell, L.H. (1998) The Saccharomyces cerevisiae RADS,RAD17, RAD24 and MEC3 Genes Are Required for Tolerating Irreparable, Ultraviolet-Induced DNA Damage. Genetics 150: 75-93.
Pegg, A.E. & Byers, T.L. (1992) Repair of DNA containing 06-alkylguanine. FASEB /. 6: 2302-2310.
Peterson, K.R. & Mount, D.W. (1987) Differential repression of SOS genes by unstable LexA41 (Tsl-1) protein causes a 'split-phenotype' in Escherichia coli K-12. J. Mol. Biol. 193: 27-40.
Petukhova, G., Stratton, S. & Sung, P. (1998) Catdysis of homologous DNA pairing by yeast Rad51 and Rad54 proteins. Nature 393: 91-94.
Prakash, L. (1974) Lack of chernically induced mutation in repair-deficient mutants of yeast. Genetics 78: 1101-1118.
Prakash, L. & Prakash, S. (1977) Isolation and Characterization of MMS- Sensitive Mutants of Saccharomyces cerevisiae. Genetics 86: 33-55. Prakash, L. (1981) Characterization of Postreplication Repair in Saccharomyces cerezisiae and Effects of rad6, radlt?, rev3 and rad52 Mutations. Mol. Gen. Genet. 184: 471-478.
Prakash, S., Sung, P. & Prakash, L. (1993) DNA repair genes and proteins of Saccharomyces cerevisiae. An nu. Rev. Genet. 27: 33-70.
Rabkin, S.D., Moore, P.D. & Strauss, B.S. (1983) In vitro bypass of W-induced lesions by Escherichia coli DNA polymerase 1: specificity of nucleotide incorporation. Proc. Natl. Acad. Sci. U S A 80: 1541-1545.
Resnidc, M.A., Boyce, J. & Cox, B. (1981) Postreplication Repair in Saccharomyces cerevisine. 1. Bacteriol. 146: 285-290.
Robzyk, K., Recht, J. & Osley, M (2000) Rad6-Dependent Ubiquitination of Histone H2B in Yeast. Science 287: 501-504.
Roca, A.I. & Cox, M.M. (1997) RecA protein: structure, fundon, and role in recombinational DNA repair. Prog. Nucleic Acids Res. Mol. Biol. 56: 129- 223.
Rodarte-Ramon, US.& Mortimer, R.K. (1972) Radia tion-induced recombination in Saccharomyces: isolation and genetic study of recombination defiaent mutants. Radia t. Res. 49: 133-147.
Rong, L., Palladino, F., Aguilera, A., & Klein, H. (1991) The Hyper-Gene Conversion hpr5-l Mutation of Sacchnrornyces cermisiae is an Allele of the SRS2/ RADH Gene. Genetics 127: 75-85.
Rong, L. & Klein, H. (1993) Purification and Charadenzation of the SRSZ DNA Helicase of the Yeast Saccharomyces cerevisiae. J. BioZ. Chem. 268: 1252- 1259.
Rose, M.D., Novick, P., Thomas, J.H., Botstein, D. & Fi&, G.R. (1987) A Smcharomyces cerevisiae genomic plasmid bank based on a centromere- containing shuttle vector. Gme 60: 237-243.
Rothofsky, M.L., & Lin, S.L. (1997) CROC-1 encodes a protein which mediates transcriptional activation of the human FOS promoter. Gene 195: 141- 149.
Rupp, W.D., Wilde, C.E., Reno, D.L. & Howard-Flanders, P. (1971) Exchanges Between DNA Strands in Ultraviolet-irradiated Escherichia coli. J. Mol. Biol. 61: 25-44.
Rymond, B.C. & Rosbash, M. (1992) Yeast Pre-mRNA Splibg, pp. 143-192, in The Molecular and Cellular Biology of the Yeast Saccharomyces, E.W. Jones, J.R. Pringle & J.R. Broach, eds. (Cold Spnng Harbor Laboratory Press, Cold Spring Harbor, NY). Sagher, D. & Strauss, B. (1983) Insertion of Nudeotides Opposite Apurùiic/Apyrimidinic Sites in Deoxribonudeic Acid during in Vitro Synthesis: Uniqueness of Adenine Nucleotides. Biochemistry 22: 4518- 4526.
Sancho, E., Vila, M.R., Sanchez-Pulido, L-, Lozano, J.J., Paciucci, R., Nadal, M., Fox, M., Harvey, C., Bercovich, B., Loukili, N., Ciechanover, A., Lin, S.L., Sanz, F., Estivill, X., Valenaa, A. & Thomson, T.M. (1998) Role of UEV-1, an inactive variant of the E2 ubiquitin-conjugating enzymes, in in vitro differentiation and ce11 cycle behavior of HT-29-M6 intestinal rnucosecretory cells. Mol. Cell. Biol. 18: 576-589.
SandelI, L.L. & Zakian, V.A. (1993) Loss of a yeast telomere: arrest, recovery, and chromosome loss. Ce11 75: 729-739.
Sanger, E., Nidden, S. & Coulson, A.R. (1977) DNA sequencing with chain- termination inhibitors. Pm.Natl. Acad. Sci. USA 74: 5463-5467.
Sarasin, A. & Hanawalt, P.C. (1980) Replication of Ultraviolet-irradiated Simian Virus 40 in Monkey Kidney Cells. J. Mol. Biol. 138: 299-319.
Sassanfar, M. & Roberts, J.W. (1990) Nature of the SOS-inducing signal in Escherichia coli: the involvement of DNA replication. J. Mol. Biol. 212: 79- 96.
Sassanfar, M., Dosanjh, M.K., Essignmann, J.M. & Samson, L. (1991) Relative efficiencies of the bacterial, yeast, and human DNA methyltransferases for the repair of 06-methylguanine and 04-methyithymine. Suggestive evidence for OPmethylthymine repair by eukaryotic methyltransferases. J. Biol. Chem. 266: 2767-2771.
Schiestl, R.H., Prakash, S. & Prakash, L. (1990) The SRS2 Suppressor of rad6 Mutations of Saccharomyces cerevisiae Acts by Channeling DNA Lesions into the RAD52 DNA Repair Pathway. Genetics 124: 817-831.
Wd,D. (1995) Suppression of a New Allele of the Yeast RAD52 Gene by Overexpression of RAD51, Mutations in sn2 and ccr4, or Mating-Type Heterozygosity. Genetics 140: 115-127.
Schnarr, M., Oertel-Buchheit P., Kâwmaier, M. & Granger-Schnarr, M. (1991) DNA binding properties of the LexA repressor. Biochimie 73: 423-431.
Çeeberg, E., Eide, L. & Bjoras, M. (1995) The base excision repair pathway. Trends Bioch. Sci. 20: 391-397. Senturker, S., Auffret van der Kemp, P., You, H.J., Doetsch, P.W., Dizdaroglu, M. & Boiteux, S. (1998) Substrate specificities of the Ntgl and Ntg2 proteins of Saccharomyces cerevisiae for oxidized DNA bases are not identical. Nucleic Acids Res. 26: 5270-5276. Sherman, F., Fink, G.R. & Hicks, J. (1983) Methods in Yeast Genetics (Cold Spring Harbor Lab. Press, Plainview, NY)
Shinohara, A. & Ogawa, T. (1995) Homologous recombination and the roles of double-strand breaks. Trends Bioch. Sci. 2: 387-391.
Snow, R. (1968) Recombination in ultraviolet-sensitive strains of Saccharomyces cerevisiae. Mut. Res. 6: 409-418.
Spence, J., Sadis, S., Haas, A.L. & Finley, D. (1995) A Ubiquitin Mutant with Specific Defeds in DNA Repair and Multiubiquitination. Mol. Cell. Biol. 15: 1265-1273.
Sugimoto, K., Sakamoto, Y., Talcahashi, O. & Matsumoto, K. (1995) HYS2, an essential gene required for DNA replication in Saccharomyces cerevisine. Nucleic Acids Res. 23: 34933500.
P., Prakash, S. & Prakash, L. (1988) The RAD6 protein of Saccharomyces cerevisiae polyubiquitinates histones, and its acidic domain mediates this activity. Genes Dm. 2: 1476-1485.
P., Prakash, S. & Prakash, L. (1990) Mutation of cysteine-88 in the Sacchnromyces cerevisiae RAD6 protein abolishes its ubiquitin-conjugating activity and its various biological functions. Proc. Natl. Acad. Sei LISA 87: 2695-2699-
P., Berleth, E., Pickart, C., Prakash, S. & Prakash, L. (1991) Yeast RAD6 encoded ubiquitin conjugating enzyme mediates protein degradation dependent on the N-end-recognizing E3 enzyme. EMBO 10: 2187-2193.
P. (1997) Yeast Rad55 and Rad57 proteins form a heterodimer that functions with replication protein A to promote DNA strand exchange by Rad51 recombinase. Genes Dev. 11: 111 1-1121.
Swanson, R.L., Morey, N.J., Doetsch, P.W. & Jinks-Robertson, S. (1999) Overlapping Specificities of Base Excision Repair, Nucleotide Excision Repair, Recombination, and Translesional Synthesis Pathways for DNA Base Darnage in Saccharomyces cerevisiae. Mol. Cell. Biol. 19: 2929-2935.
Tang, M., Shen, X., Frank, E.G., OfDomell,M., Woodgate, R. & Goodman MF. UmuD1(2)Cis an error-prone DNA polyrnerase, Escherichia coli pol V. Proc. Natl. Acad. Sci. U S A 96: 8919-8924.
Toczyski, D.P., Galgoczy, DJ. & Hartwell, L.H. (1997) CDC5 and CKII control adaptation to the yeast DNA damage checkpoint. Ce11 90: 1097-1106.
Torres-Ramos, C.A., Yoder, B.L., Burgers, P.M.J., Prakash, S., & Prakash, L. (1996) Requirement of proliferating ce11 nudear antigen in RAD6- dependent postreplication DNA repair. Proc. Natl. Acad. Sci. USA 93: 9676-9681. Torres-Ramos, C.A., Prakash, S., & Prakash, L. (1997) Requirement of Yeast DNA Polymerase 6 in Postreplication Repair of UV-damaged DNA. 1. Biol. Chem. 272: 2544525448.
Varshavsky, A. (1997) The N-end depathway of protein degradation. Genes to Cells 2: 13-28.
Voet, D. & Voet, J-G. (1990) Biochemisty Uohn Wiley & Sons, Toronfo).
Von Borstel, R.C. (1978) Measuring spontaneous mutation rates in yeast. Methods Ce11 Biol. 20: 1-24.
Wagner, J., Gruz, P., Kim, S., Yamada, M., Matsui, K., Fuchs, R.P.P. & Nohmi, T. (1999) The dinB Gene Encodes a Novel E. coli DNA Polymerase, DNA Pol IV,Involved in Mutagenesis. Mol. Ce11 4: 281-286.
Walker, G.C. (1995) SOSregulated proteins in translesional DNA synthesis and mutagenesis. Trends Bioch. Sri. 20: 41 6-420.
Wang, 2. & Rossrnan, T.G. (1994) Isolation of DNA fragments from agarose gel by centrifugation. Nucleic Acids Res. 22: 2862-2863.
Washington, M.T., Johnson, R.E., Prakash, S. & Prakash, L. (1999) Fidelity and Processivity of Saccharomyces cerevisiae DNA Polymerase v. 1. Biol. Chem. 274: 36835-36838.
Watkins, J.F., Sung, P., Prakash, S. & Prakash, L. (1993) The extremely conserved amino terminus of RAD6 ubiquitin-conjugating enzyme is essential for amino-end de-dependent protein degradation. Genes. Dev. 7: 50-261.
Weinert, T. (1998) DNA darnage checkpoints update: getting molecular. Curr. Opin. Genet. Deuel. 8: 185-193.
Williamson, M.S., Game, J.C. & Fogel, S. (1985) Meiotic gene conversion mutants in Saccharomyces cermisiae. Genetics 110: 609-646.
Wilson, D.M. & Thompson, L.H. (1997) Life without DNA repair. Proc. Natl. Acad. Sci. USA 94: 12754-12757.
Wyatt, M.D., Allan, J.M., Lau, A.Y., Ellenberger, T.E. & Samson, L.D. (1999) 3- Methyladenine DNA glycosylases: structure, function, and biological importance. BioEssays 21: 668-676.
Xiao, W. & Samson, L. (1993) In vivo evidence for endogenous DNA alkylation darnage as a source of spontaneous mutation in eukaryotic cells. Proc. Natl. Acad. Sci. USA 90: 2117-2121. Xiao, W., Chow, B.L. & Rathgeber, L. (1996) The repair of DNA methylation damage in Saccharomyces cerevisiae. Curr. Genet. 30: 461-468.
Xiao, W ., Lin, S.L., Broomfield, S., Chow, B.L. & Wei, Y.F. (1998) The products of the yeast MMS2 and two human homologs (hW2and CROC-1) define a structuralIy and functionally conserved Ubc-like protein family. Nucleic Acids Res. 26: 3908-3914.
Zavitz, K.H. & Marians, K.J. (1991) Dissecting the functional role of PriA protein-catalysed primosome assembly in Escherichia coli DNA replication. Mol. Microbiol. 5: 2869-273.
Zou, H. & Rothstein, R. (1997) Holliday Junctions Accumulate in Replication Mutants via a RecA Homolog-Independent Mechanism. Ce11 90: 87-96.