r '.' a ,._ .‘l '4 c o .‘ 7'1 n . {. ~31.“ , J ... ,. ‘5‘. .. < ;: ‘4 5 23v. ”1:. . 1" , tmf‘x" .. ‘ u 0 ‘éx' c.- ”.1.” . ‘n.’ I. .s ml .» .‘V‘ w w n ‘ E ,. , ‘ ‘31.:3‘13‘ “ ‘-4.x w. .l a” , W... 7.5; .3-..;.:. a, .1 .. V,. ~¢v J 5U": u‘ . (.1. x 1- \‘m1 «3.; fi' ;4’.« f»; Np.“ Jr}; "in' '1 (4.3.7 CNTK 1 mg??? avg“ ;: g. a... 5 (fix- 2‘2 1..‘~1 {as 4‘ ,4r_ Ju“' ‘1 mm :, nun“ ”h”, . 4. Q 55;}?qu «:g t 5" . ,. . .. .. n ., Jinn; { A u, . n. , um fl": 1. Hum“. -, , .« .,....,.._, ,u ”Wu". . J. u-u -~ v ‘2"...utu .L'I'. .. 4.3. r. .~oo..¢', ”..~~......, «gum... . - m '2 u :u' up. {2.3. ‘ a". n... , " , - . -'\V' o " . . . . ,. uJ. win}... '2‘- 'I. ‘v 'wm , ~ row~ - , ‘7; or" ' '1, le ...’r"" . .. up! .3" < u @515 NSIVER SIIBTYL llllllllllllllllllHHlllll|| llllll 31293 00900 HIIIHZIHIHI This is to certify that the dissertation entitled 3'—5' EXONUCLEASE IN DROSOPHILA MITOCHONDRIAL DNA POLYMERASE: KINETIC AND MECHANISTIC STUDIES presented by MATTHEW W. OLSON has been accepted towards fulfillment of the requirements for We 5.1mm: Major professor Date W 3: :99: MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 —.¥ r ——~ LIBRARY mcllllan State l University "u A W PLACE IN RETURN BOX to remove thle checkout from your record. TO AVOID FINES return on or before dale due. DATE DUE DATE DUE DATE DUE 7W» MSU le An Affirmative Action/Equal Opportunlty Institution Mans-o1 ABSTRACT 3'-5' EXONUCLEASB IN DROSOPHILA MITOCHONDRIAL DNA POLYMERASE: KINETIC AND MECHANISTIC STUDIES by Matthew Wayne Olson The mitochondrial DNA polymerase from Drosophila melanogaster embryos lacks dNTP turnover activity. However, a potent 3'-5' exonuclease activity can be detected by a specific assay in which the exonuclease hydrolyzes mispaired nucleotides at the 3'-termini of primed synthetic and natural DNA templates. The 3'-5' exonuclease copurifies quantitatively with DNA polymerase y and cosediments with the nearly homogeneous enzyme under native conditions. The rate of mispair excision is 14.5- to 34-fold greater than the rate of excision of a correctly paired nucleotide on a synthetic DNA template-primer. The excision of a 3'-terminal mispair is at least 14-fold more efficient than the excision of a 3'-terminal base pair on a natural DNA primer-template. The ratio of DNA polymerase activity to 3'-5' exonuclease activity varies depending upon the template-primer used from approximately 3:1 to 1700:1. The 3'-5' exonuclease appears to be functionally coordinated with the DNA polymerase. First, the monovalent cation optima for each activity are nearly identical on two different DNA substrates. Second, the divalent metal cation optima for the two enzyme activities are essentially the same on singly-primed M13 DNA. Third, the affinities of the DNA polymerase and 3'-5' exonuclease for singly-primed phage M13 DNA are very similar. Fourth, the 3'-5' exonuclease is functional under conditions of DNA synthesis, and is capable of hydrolyzing 3'-terminal mispairs prior to primer extension, providing strong evidence for a functional coordination of the two enzyme activities. The Drosophila Y'polymerase will not extend a mispair when in a "stationary" state even when subjected to conditions of a large next nucleotide bias. Instead, under such conditions, 3'- terminal mispairs are hydrolyzed quantitatively by the 3'-5' exonuclease over the reaction time course. During DNA synthesis by the y’polymerase in the polymerization mode, limited misincorporation and subsequent mispair extension do occur. Here it appears that misincorporation and not primer extension of a mispair is rate limiting. Preliminary evidence suggests the mechanism of template-primer transfer from the 3'-5' exonuclease active site to DNA polymerase active site following mispair excision, is intermolecular. DNA polymerase and 3'-5' exonuclease inhibition studies, using potent and specific polyclonal antiserum to the Drosophila y’polymerase, demonstrate a physical association of the two enzyme activities. Further, immunoprecipitation of a Drosophila embryo homogenate results in the detection of 125 kDa and 35 kDa polypeptides, indicating that the y'polymerase, as purified to near-homogeneity is intact. Finally, partial chemical digestion of the purified a— and B—subunits strongly suggests that the two subunits are proteolytically distinct. Whatever you do, whether in word or deed, do it in the name of the Lord Jesus giving thanks to God the Father through Him. The Apostle Paul in Col.3:17 ACKNOWLEDGMENTS First and foremost I thank Laurie S. Kaguni for her time, patience and guidance as an excellent thesis advisor: I acknowledge her great contribution to the work described herein. I express my gratitude for the scientific skills I have gained under her watchful eye. I also thank my wife, Jennifer, for her encouragement in times of deep stress and her tolerance of my way of thought. I thank Doug and Andrea Cress for their input both in and outside the laboratory and for being great friends over the past five years. I thank Andrea Von Tom, my lab mate, for painstakingly correcting many versions of this dissertation and for the many positive critisms of my scientific research. I would like to thank Dr. Zach Burton for his time, and for thought provoking conversations with regard to this work and for being a friend. I also like to thank Drs. Tom Friedman, William Smith, Tom Deits, and Zach Burton for their service as members of my guidance committee. I thank Jon Kaguni for his insightful contributions to my graduate work at the Kaguni Laboratory group meetings. I also acknowledge the encouragement and assistance I have received from members of the Kaguni laboratories: Andrea VonTom, Ted Hupp, Catherine Wernette, Dave Lewis, Carol Farr, Kevin Carr, Cindy Petersen, Jarek Marzsalek, Rhod Elder, Carla Margulies, Doug Su Hwang, and Qingping Wang. Finally, I thank my parents for their years of love and support and for providing a firm foundation to build upon and eventually succeed. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES LI ST OE ABBREVIATIONS CHAPTER I: LITERATURE REVIEW The Mitochondrion Mitchondrial Biogenesis Mitochondrial DNA DNA Replication Prokaryotic DNA Replication Eukaryotic DNA Replication Mitochondrial DNA Replication Prokaryotic DNA Polymerases Eukaryotic DNA Polymerases 3'-5' Proofreading Exonucleases Structure-Function Studies of DNA Polymerases CHAPTER II: EXPERIMENTAL PROCEDURES MATERIALS Nucleotides and nucleic acids Enzymes and protein Standards Chemicals xi xii XV 10 18 19 25 31 36 41 49 52 5'7 5'7 57 58 58 METHODS DNA polymerase Y assay Deoxynucleotide turnover assay 3'-5' Exonuclease assay on synthetic DNA substrates Preparation of 3'-[3H]- labeled DNase I-activated calf thymus DNA for exonuclease assay Preparation of 5'-[32P]-labeled substrates for product analysis after exonuclease assay 3'-5' Exonuclease assay on 3'-[3H]- labeled DNase I-activated calf thymus DNA 3'-5' Exonuclease assay on natural DNA and product analysis by denaturing gel electrophoresis Steady-state kinetic analysis of template-primer affinity of the 3'-5' exonuclease Nucleoside monophosphate inhibition of the 3'-5' exonuclease . . . . Primer extension analysis in the presence or absence of a 3'-terminal mispair Trapping experiment to detect transfer from the 3'-5' exonuclease site to the DNA polymerase site Preparation of embryo extract Preparation of a high speed supernatant fraction Protein determinations Preparation of antiserum DNA polymerase and 3'-5' exonuclease inhibition studies Protein gel electrophoresis, protein transfer, and immunoblotting . . Glycerol gradient sedimentation under native conditions . . . . ‘vii (II \D U] \i) 60 61 61 62 62 63 63 64 65 65 65 65 66 66 67 Glycerol gradient sedimentation under denaturing conditions Sephacryl 8-200 gel filtration chromatography under denaturing conditions Elution of proteins from sodium dodecyl sulfate- polyacrylamide gels Immunoprecipitation of the Drosophila y polymerase Chemical cleavage of the Drosophila y'polymerase CHAPTER III : A MISMATCH SPECIFIC 3 ' -5 ' EXONUCLEASE ASSOCIATED WITH THE MITOCHONDRIAL DNA POLYMERASE PROM DROS OPE I LA EMBRYOS INTRODUCTION RESULTS Drosophila DNA polymerase y'does not catalyze detectable dNTP turnover during DNA polymerization . . . A 3'-5' Exonuclease activity associated with Drosophila mitochondrial DNA polymerase excises 3'-terminal mismatched nucleotides from synthetic DNA Quantitative copurification of 3'-5' exonuclease and DNA polymerase Cosedimentation of 3'-S' exonuclease and DNA polymerase Drosophila DNA polymerase Y exhibits exonucleolytic editing under conditions of nucleotide polymerization DISCUSSION CHAPTER IV: COORDINATION Ol' 3'-5' EXONUCLEASE AND DNA POLYMERASE FUNCTION IN DROSOPHILA MITOCHONDRIAL DNA POLYMERASE viii 68 69 69 7O 71 72 72 74 74 74 76 81 81 90 94 INTRODUCTION . . . . . . . . 94 RESULTS . . . . . . . . . 98 The monovalent cation optima for DNA polymerase and 3'-5' exonuclease activities on singly-primed M13 DNA and DNase 1-activated calf thymus DNA are similar. . . . . 98 Divalent cation optima of the DNA polymerase and the 3'-S' exonuclease are similar. . . 103 Primer-template affinity of the 3'-5' exonuclease. . . . . . . 106 Inhibition of the 3'-5' exonuclease by S’AMP . 106 Differential excision of 3'-terminal mispairs . lll Specificity of mispair excision . . . 112 Exonucleolytic editing during DNA synthesis and next nucleotide effect . . . . 117 Mechanism of template-primer transfer from the 3'-S' exonuclease to 5'-3' DNA polymerase site . . . . . . 125 DISCUSSION . . . . . . . . 131 CHAPTER V: STRUCTURE-FUNCTION STUDIES OF THE DROSOPHILA MELANOGASTER MITOCHONDRIAL DNA POLYMERASE . . . . . . 1 4 0 INTRODUCTION . . . . . . . . 140 RESULTS . . . . . . . . . 144 Specificity of the antiserum developed against the Drosophila y’polymerase . . . . 144 Inhibition of DNA polymerase and 3'-5' exonuclease by anti-DNA polymerase 7 serum . 144 .ix Dissociation and separation of the subunits of Drosophila y polymerase subunits Immunoprecipitation of DNA polymerase Y from crude extracts Partial proteolysis of the purified 125 kDa and 35 kDa subunits of DNA polymerase Y . DISCUSSION FUTURE STUDIES REFERENCES 165 169 172 180 185 List of Tables CHAPTER. III Table 1. Measurements of dNTP turnover catalyzed by D. melanogaster DNA polymerase Y'and by E. coli DNA polymerase I and DNA polymerase III holoenzyme. . 75 Table 2. 3'-5' Exonuclease activity of Drosophila DNA polymerase Y . . . . . . . . 77 CHAPTER. IV Table 3. Template-primer affinity of the 3'-5' exonuclease and DNA polymerase associated with the Drosophila y'polymerase for singly-primed DNA. . . 107 Table 4. Inhibition of the 3'-5' exonuclease activity associated with the Drosophila y polymerase by adenosine 5' monophosphate (AMP). . . 109 Table 5. The Ki for AMP in mispair excision by the 3'-5' exonuclease associated with Drosophila y’polymerase: Dixon and Lineweaver-Burk analyses . . 111 CHAPTER.‘V Table 6. Recovery of Drosophila DNA polymerase Y and E. coli DNA polymerase I following SDS-PAGE, elution from a gel slice, acetone precipitation, and resuspension in and dilution from 6 M guanidine hydrochloride . . . . . . . . . 164 xi List of Figures Chapter' I Figure 1. Hypothetical model of protein import into mitochondria . . . . Figure 2. Structure and genetic map of vertebrate mitochondrial DNA . . . Figure 3. Structure and genetic map of Drosophila mitochondrial DNA Figure 4. Scheme for enzymes operating at a replication fork of E. coli Figure 5. Scheme for concurrent replication of leading and lagging strands by and asymmetric dimeric DNA polymerase associated with a primosome and a helicase. . . . . . Figure 6. Model of the SV40 DNA replication fork and the enzymes involved. . . . . Figure 7. Replication model for monomeric mouse mitochondrial DNA. Figure 8. A scheme for the replication of Drosophila mitochondrial DNA. . . . . . . CHAPTER. III Figure 9. Copurification of the 3'-5' exonuclease and DNA polymerase Figure 10. Cosedimentation of 3'-5' exonuclease and DNA polymerase Figure 11. Proofreading DNA synthesis on M13 DNA by DNA polymerase 7 Figure 12. Time course of utilization of paired versus mispaired primers on M13 DNA . xii. 13 15 22 24 28 33 38 80 83 86 89 CHAPTER. IV Figure 13. Dependence of DNA polymerase and 3'-5' exonuclease activities on monovalent cation concentration in the presence of a singly-primed DNA substrate. Figure 14. Dependence of DNA polymerase and 3'—5' exonuclease activities on monovalent cation concentration in the presence of a DNA substrate with high primer density. Figure 15. Dependence of DNA polymerase and 3'-5' exonuclease activities on divalent cation concentration in the presence of a singly-primed DNA substrate. Figure 16. Time course of hydrolysis of 3'- terminal mispairs Figure 17. Examination of mispair specificity of the 3'-5' exonuclease associated with y’polymerase. Figure 18. Template-primer utilization by Drosophila DNA polymerase 7 -- next nucleotide effect under "stationary conditions." Figure 19. Template-primer utilization by Drosophila DNA polymerase 7 -- next nucleotide effect under "synthesis conditions." Figure 20. Quantitation of template-primer utilization by Drosophila Y'polymerase Figure 21. Transfer of the template-primer from the 3'-5' exonuclease site to the 5'-3' DNA polymerase site of Drosophila Pol y CHAPTER.‘V Figure 22. Reactivity of rabbit antiserum against DNA polymerase 7 Figure 23. Immunoprecipitation of DNA polymerase y xiii 100 102 105 114 116 119 121 124 128 146 148 Figure 24. Inhibition of DNA polymerase Y’by rabbit antiserum . . . . . . . 150 Figure 25. Cosedimentation of DNA polymerase and 3'-5' exonuclease in the presence of ethylene glycol . . . . . . . 154 Figure 26. Gel filtration of Drosophila DNA polymerase Y in guanidine hydrochloride . . 157,158,159 Figure 27. Glycerol gradient sedimentation of Drosophila y’polymerase in the presence of guanidine hydrochloride . . . . . . 162 Figure 28. Immunoprecipitation of Drosophila DNA polymerase y from crude fractions . . . . 167 Figure 29. Partial proteolytic digestion of the a- and fl-subunits of Drosophila ‘Y polymerase . . 171 xiv' A+T BCA hp BrdUMP BSA CPNI CSB DI DHFR D-Loop dNTP DTT 3. coli GIP Gu-RCl HSP hop HSSB R-strand HSV HTLVI IgG IMM kl) LIST OF ABBREVIATIONS deoxyadenylate and thymidylate bovine carbonic anhydrase base pair 5'-bromo deoxyuridine monophosphate bovine serum albumin counts per minute conserved sequence blocks dalton dihydrofolate reductase displacement loop deoxynucleoside triphosphate dithiothreitol Escherichia coli general insertion protein guanidine hydrochloride heavy strand promoter heat shock protein human single-standed DNA binding protein heavy strand herpes simplex virus human T-cell leukemia virus 1 immunoglobulin G inner mitochondrial membrane kilobase XV kDa. ZLSP L-atrand MOW! MPP mtDNA NCS NEH! NMP nt NTP 8-N3-dATP On 0L OMM PBS PCNA PEP Pol (x Pol l3 Pol 8 Pol 8 Pol y Pol I kilodalton light strand promoter light strand mitochondrial outer membrane protein mitochondrial processing peptidase mitochondrial DNA N-chlorosuccinimide N-ethylmaleimide nucleoside monophosphate nucleotide nucleoside triphosphate 8-azido deoxyadenosine triphosphate heavy strand origin light strand origin outer mitochondrial membrane phosphate buffered saline proliferating cell nuclear antigen processing enhancing protein DNA polymerase a DNA polymerase B DNA polymerase 5 DNA polymerase 8 DNA polymerase y E. coli DNA polymerase I xvi. Pol I n' E. coli DNA polymerase I Klenow fragment Pol III E. coli DNA polymerase III Pol III HE E. coli DNA polymerase III holoenzyme PPZA protein phosphatase 2A RFA replication factor A RFC replication factor C RNase MRP RNA processing endoribonuclease SSB single-stranded DNA binding protein SV40 simian virus 40 T38 tris buffered saline TBST tris buffered saline + 0.05% Tween 20 Tx-IOO triton X-lOO xvii CHAPTER I LITERATURE REVIEN THE MI TOCHONDRION The mitochondrion is a subcellular organelle found in nearly all eukaryotic organisms and is the only organelle in animal cells which contains its own DNA (1,2,3). The mitochondrion is the center of energy metabolism in the cell, as it is responsible for ATP synthesis. Oxidative phosphorylation and electron transport are arguably the most important functions of mitochondria in terms of cellular energy production (4). The mitochondrion is surrounded by a double membrane and is 1-2 um long and 0.5-1 um.wide. The inner and outer membranes serve distinct roles; thus, it is not surprising that their lipid and protein content differ (2). The outer mitochondrial membrane (OHM) contains a substantial number of enzymes. However, the enzymes present in this membrane do not constitute an integrated metabolic pathway. Consequently, the compositional data do not give us any clear idea about the function of the OHM (5). The mitochondrial outer membrane does contain a number of enzymes involved in phospholipid biosynthesis. In addition, the outer membrane contains many copies of a channel forming protein(s) allowing small molecules and targeted proteins free access to the intermembrane space. The inner mitochondrial membrane (IMM) contains those proteins utilized in electron transport and the enzymes involved in oxidative phosphorlylation. The major phospholipids of mitochondria comprising the mitochondrial membranes are phosphatidyl choline, phosphatidyl ethanolamine and cardiolipin. The lipid content of the fractionated inner and outer membranes differ as demonstrated by the lower density of the OMM in velocity sedimentation experiments (5). The IMM contains an unusually high content of cardiolipin and is relatively impermeable even to small ions (6). In rat liver mitochondria, the IMM contains 21% of the mitochondrial proteins. Nearly all of the remaining proteins and the mitochondrial DNA are located in the mitochondrial matrix (2). Of particular significance are the enzymes involved in the tricarboxcylic acid cycle, the fatty acid oxidation cycle and the pyruvate dehydrogenase complex. In addition, the matrix contains transfer RNAs, ribosomal RNAs, the DNA polymerase and the RNA polymerase and other components of the mitochondrial DNA replicational, transcriptional, and protein translational machinery necessary for mitochondrial biogenesis. MITOCHONDRIAL BIOGENESIS Mitochondria contain some 300 to 400 different proteins. With few exceptions these proteins are specified by nuclear genes and synthesized in the cytoplasm. The import of these proteins and the expression of those proteins encoded by the mitochondrial DNA are necessary for replication of mitochondrial DNA and subsequent mitochondrial biogenesis, respectively. Defining the roles of mitochondrial and nuclear DNA in directing the biogenesis of respiratory-competent mitochondria prompts three questions. What is the degree of genetic autonomy of mitochondria? How is the expression of mitochondrial DNA regulated by nuclear gene products? How are the cytoplasmically synthesized proteins transported into the mitochondria? The ability of mitochondria to divide and fuse is well established from direct microscopic observations. However, the details of this process are not well understood (2). Mitochondria occupy a unique position, in terms of their genetic autonomy, among cellular organelles because of their possession of a separate genome and all the enzymatic machinery for translating the genetic information into functional proteins. Earlier observations that mitochondria synthesize certain constituent polypeptides of cytochrome oxidase (7), coenzyme 0H2 reductase (8), and ATPase (9) suggested that these proteins were encoded by mitochondrial DNA (mtDNA). This has been borne out by genetic analysis of yeast mtDNA (10) and subsequent sequencing of the mitochondrial genome from fungal (11), animal (12,13,14), and plant Sources (15). In Saccharomyces cerevisiae it is possible to induce point mutations or small deletions in mtDNA that result in a respiratory deficiency, mit'. Such mutants retain a fully functional system of mitochondrial protein synthesis (16). These mutant phenotypes are the result of lesions in structural components of three IMM enzymes. The genetic automomy of mitochondria is better illustrated by a second set of mutations which have been termed syn ‘. The disruption of these genes results in the disruption 4 mitochondrial protein synthesis (17). They include mutations in the 15S and 218 rRNAs, 25 tRNAs and a single protein subunit of mitochondrial ribosomes. These mutants are also respiratory incompetent since they cannot synthesize the normal complement of mitochondrially encoded respiratory and ATPase proteins. Therefore, respiratory competent mitochondria require intact protein translation machinery as well as an intact genetic code. It has been apparent that despite the importance of mtDNA for the maintenance and propagation of respiratory functional mitochondria, the most fundamental processes involved in biogenesis of the organelle are dependent on the expression of genes located in nuclear DNA. The very fact that genetic information is distributed among two spatially separated compartments implies the existence of mechanisms for ensuring coordinate expression of the proteins encoded in the two genomes. This is most clearly illustrated by the long-standing observation that p° mutants of yeast lacking intact mtDNA nonetheless have organelles morphologically and functionally related to mitochondria. Such respiratory-deficient mitochondria have most of the enzymatic capabilities of wild- type mitochondria except for the few respiratory and ATPase proteins encoded in mtDNA (18). Classes of nuclear genes that code for products with a direct function in mitochondrial respiration, oxidative phosphorylation or oxidative metabolism of mitochondria have recently been examined. One such class is the PET genes, which are nuclear genes whose expression is required for the morphogenesis of respiratory-competent mitochondria. These genes may code for products that have a direct function in mitochondrial respiration and oxidative phosphorylation or they may affect oxidative metabolism of mitochondria indirectly. For example, a gene coding a mitochondrial ribosomal protein qualifies as PET since defective mitochondrial protein synthesis would lead to a respiratory deficient phenotype. The occurrence of separate promoters for different genes in yeast mtDNA affords the possibility of individual regulation of gene expression. This could be achieved by means of trans acting protein factors encoded in nuclear DNA. The PET 111, 122, 54 and 494 gene products have been shown to specifically activate the translation of yeast mitochondrial mRNA encoding cytochrome c oxidase subunit III (coxIII), a protein required for cellular respiration (19,20,21). The specific activation of coxIII translation by the PET genes appears to occur by the action of these gene products at a site or sites in a region of the 5' untranslated leader sequence upstream of the initiation codon (21). There are four well-documented examples of mitochondrial enzymes with subunit polypeptides derived from both nuclear and mitochondrial genes (22). These enzymes include the two respiratory complexes of cytochrome oxidase, coenzyme 0H2- cytochrome 0 reductase, the oligomycin sensitive ATPase and mitochondrial ribosomes. The inverse question, of whether mitochondria affect the expression of nuclear DNA, has also been examined. The answer appears to be negative since mitochondria 6 of wild-type and of p0 mutants have essentially identical protein composition, except for the proteins translated on mitochondrial ribosomes (18). This indicates a lack of mitochondrial involvement in regulating either transcription of PET genes or translation of cytoplasmic mRNAs. Nuclear encoded enzymes and proteins play an important role in the replication and transcription of mtDNA, in processing of precursor RNAs and in translation of mitochondrial messages. The DNA and RNA polymerases and virtually all the proteins that constitute the replicational and transcriptional machinery of mitochondria are encoded by nuclear genes. Thus, the maintenance of a wild type mitochondrial genome and a respiratory competent cell is dependent upon the import of such proteins into the organelle. The exact mechanism of protein import and refolding has not yet been completely elucidated; however, models have been proposed. One model describing the import of nuclear encoded mitochondrial proteins is presented in Figure 1. Precursor proteins synthesized in the cytosol are recognized by receptors on the mitochondrial surface via presequences on the polypeptide (23-25). Mitochondrial presequences are amino-terminal, at least 70 residues in length, target polypeptides to the mitochondria, and are removed by one or two proteolytic steps in the mitochondria. The OMM receptor protein is referred to as MOM 19 and is a mitochondrial outer membrane protein 19,000 daltons in molecular weight. The precursors are inserted into Figure l. Hypothetical model of protein transport into mitochondria. MOM 19, a mitochondrial outermembrane protein of 19 kDa; MOM 72, a mitochondrial outermembrane protein of 72 kDa; GIP, general insertion protein; PEP, processing enhancing protein; MPP mitochondrial processing pepsidase; hsp60, a heat- shock protein of 60 kDa. Taken from Pfanner N. and Neupert, W. (1990) Ann. Rev. Biochem. 59, p.341. Outer #4 membrane Inner membrane Matrix, Precursor proteins with presequence (and. other precursor proteins) “UNIS GIP Contact sites PEP Inner r:— membrane, Intermembrane space MPP RspGO Figure 1 Possible interaction with cytosolic cofactors Receptors on the mitochondrial surface General insertion protein in the outer membrane Electrical potential for entry into the inner membrane Proteolytic processing and (re)folding in the matrix the OMM and are translocated via the general insertion protein (GIP), presumably in an unfolded conformation. The precursor proteins are thought to be translocated through a proteinaceous channel, where the membrane potential is required for initial entrance of the "presequence" into the inner membrane (26). During and/or following translocation of the mitochondrial precursor protein into the matrix, the mitochondrial processing peptidase (MPP) removes the mitochondrial presequence (27,28). The protein is folded into an active conformation with the assistance of the processing enhancing protein (PEP) and the heat shock proteins hsp 60 and hsp 70 (27,29,30). The folding of the imported proteins and assembly of functional complexes has been proposed to occur via the assistance of "molecular chaperones" (29, 30). Molecular chaperones are defined as a family of unrelated cellular proteins that mediate the correct assembly of other polypeptides, but are not themselves components of the final structures (29,30). In terms of identification and characterization of the components required for mitochondrial protein import, stress proteins or heat shock proteins have been shown to be involved in several steps of protein transfer into the mitochondria under conditions of normal cell growth (25). These steps include maintaining the imported proteins in a translocational competent form and refolding of those proteins in the matrix. These proteins have been grouped into several families according to similarities in their primary structure (31). 10 Of particular interest are two heat shock protein families hsp 60 and hsp 70. An example of the hsp 60 family is the subunit binding protein of chloroplasts consisting of a 61 kDa a chain and a 60 kDa B chain. Both of these chains have been shown to be required for the correct assembly of rubisco in a MgATP dependent manner (32). The temperature sensitive lethal nuclear mutation in yeast described as mif 4 has been shown to encode a 60 kDa heat shock protein (33). In vitro analysis has demonstrated that yeast hsp 60 is required for the folding of recombinant dihydrofolate reductase (DHFR) in an ATP dependent manner following translocation into mitochondria (34). Evidence that hsp 70 proteins are involved in protein translocation in eukaryotic cells has been derived from studies with yeast (35). One hsp 70 protein, Ssclp, is located in the mitochondria and performs an essential cellular function in mitochondrial import of nuclear encoded proteins since disruption of SSCl results in lethality (36). MITOCHONDRIAL DNA Unlike other organelles, mitochondria are similar to the nucleus because both are surrounded by a double membrane and contain DNA (37,38). Mitochondrial DNA (mtDNA) comprises less than 1% of the total cellular DNA and the number of mitochondrial genomes per cell has been estimated to be >1000 in mouse L—cell fibroblasts and >8000 in cultured human cells (39). The density of mitochondrial organelles is uniform throughout the cell cycle. Further, mtDNA has been shown to replicate 11 during any phase of the cell cycle in mouse L cells (40). However, it has not been established whether duplication of the organelle and replication of the mitochondrial genome are coordinated. The mtDNA of metazoans resides in the matrix as a closed circular double-stranded DNA species with a genome size of 14.5 to 19.5 kilobase pairs (Figs. 2 and 3) (41). While nuclear DNA exists as a flexible chain of nucleosomes, nucleosome formation does not occur in mtDNA and neither histones nor histone-like proteins are present. However, mtDNA has been isolated in the form of DNA-protein complexes (42,43). Further, in Drosophila certain sites within a segment of the mitochondrial genome which are rich in adenine and thymine nucleotides have been shown to be resistant to crosslinking reagents (44). These studies suggest that the mtDNA is associated with protein at discrete sites. Physical studies have revealed several distinct features of mtDNA. First, the mitochondrial genome in mammals and Drosophila is arranged so that in nearly all cases one or more tRNA genes are located between the protein encoding or rRNA genes (45,46). Second, in mammalian mtDNA there is an over- representation of thymidylate residues in the second base position of codons. Because most protein coding sequences are on one strand, the duplex mammalian mtDNA is readily separated into the heavy (H-) and light (L-) strands by buoyant density gradient centrifugation (47). Figure 2. Structure and genetic map of vertebrate mitochondrial DNA. OH, origin and direction of H-strand mitochondrial DNA replication; 0L, origin and direction of L- strand mitochondrial DNA replication; D, D-loop region of vertebrate mitochondrial DNA; 125 rRNA, coding region for the small ribosomal RNA subunit; 16s rRNA, coding region for the large ribosomal subunit; NDl, NDZ, ND3, ND4, NDS, ND6, coding regions for the six subunits of the NADH dehydrogenase complex; cyt b , coding region for cytochrome b; COII, COII, COIII, coding regions for the cytochrome oxidase subunits I, II and III; A6, coding region for ATPase subunit 6; A8, coding region for ATPase subunit I.’ Capital letters represent the standard nomenclature for the tRNA encoding genes. The arrow associated with each coding region indicates the direction of transcription. Taken from Jacobs, H.T., Elliott, D.J., Math, V.B. and Farquharson, A (1988) J; MOI. Biol. 202, 185-217. 13 VERTEBRATES Figure 2 14 Figure 3. Structure and genetic map of Drosophila mitochondrial DNA. or, origin and direction of leading-strand mitochondrial DNA replication; A+T, A+T rich region of Drosophila mitochondrial DNA; 123 rRNA, coding region for the small ribosomal RNA subunit; 16s rRNA, coding region for the large ribosomal subunit; NDl, ND2, ND3, ND4, ND5, ND6, coding regions for the six subunits of the NADH dehydrogenase complex; cyt b , coding region for cytochrome b; COII, COII, COIII, coding regions for the cytochrome oxidase subunits I, II and III; A6, coding region for ATPase subunit 6, A8, coding region for ATPase subunit I. Capital letters represent the standard nomenclature for the tRNA encoding genes. The arrow associated with each coding region indicates the direction of transcription. Taken from Jacobs, H.T., Elliott, D.J., Math, V.B. and Farquharson, A (1988) J; Mol. Biol. 202, 185—217. 15 DROSOPHILA Figure 3 16 The predominant form of mammalian mtDNA is a covalently closed circle with a displacement loop (D-Loop) region at or near the origin of replication of the H-strand (48,49) (Fig. 2). The D- loop is a triple-stranded structure formed by synthesis of a third strand complementary to the L-strand and varies in length from 520-700 nucleotides. The D-loop region is bordered by the tRNAPhe and tRNAPr0 structural genes and is the only region in the mitochondrial genome of substantial length that does not contain protein encoding sequences. The function of the D-loop strands with regard to DNA replication or RNA transcription has not been defined. However, the D-loop region does contain the H-strand origin of replication, 03. Further, the in vivo transcriptional initiation sites for both the H- and L-strands map within the D-loop region (50,51). In vitro transcriptional assays using partially fractionated enzyme from human mitochondria have allowed the identification of the precise sequences required for accurate transcription initiation (52). Both promoters span a region of ~50 base pairs which are necessary and sufficient for accurate initiation of transcription in vitro (53). Therefore, it appears reasonable to assume that this area of the genome plays an important part in the control of mammalian mitochondrial gene expression. In the genus Drosophila, the mtDNA ranges in size from 15.7 to 19.5 kilobase pairs. The variability of genome size is due to the variability in the size of a single region that is rich in deoxyadenylate and thymidylate residues and is termed the A+T region. D-loop structures have not been found (54-57) (Fig. 3). 17 This region is 95% deoxyadenylate and thymidylate and was discovered by heat denaturation studies of Drosophila mtDNA replication intermediates (58-60). It varies in size.between 1.1 and 5.1 kb in the melanogaster group (61) and is approximately 1 kb in all other Drosophila species. Similar to the D-loop region, there is no evidence that the A+T rich region encodes proteins due to its low G+C content (5%), high frequency of termination codons, and lack of conserved sequences between the various species (62). Heteroduplex mapping experiments have shown that the nucleotide sequences in the A+T rich region of different Drosophila species have diverged extensively via various rearrangements (63). However, this A+T rich region is located in the identical position relative to the remainder of the mitochondrial genome in all Drosophila species examined. The function of the A+T rich region in Drosophila, like the D- loop region in mammalian systems is not known, although the origins of mtDNA replication for six species of Drosophila have been localized to this area by denaturation mapping, and analysis of restriction digests of partially replicated mtDNA molecules by electron microscopy (12). Though considerable rearrangements have occurred, the mitochondrial genome has been largely conserved throughout evolution with respect to size, shape, and gene content. In addition, the mitochondrial genomes isolated from animal sources studied thus far are eXtremely compact with little or no intergenic spaces. The mitochondrial genomes of human, bovine, mouse and some Drosophila species have been entirely sequenced 18 and the mitochondrial gene products identified (12,13,66—70). These studies demonstrated that the mtDNA encodes two rRNAs (128 and 16S) and twenty-two tRNAs necessary and sufficient for mitochondrial protein synthesis. Further, thirteen mitochondrial gene products have been identified as components of the enzyme complexes located in the IMM that function in electron transport and oxidative phosphorylation. These are the cytochrome oxidase subunits I, II an III, the NADH dehydrogenase subunits I, II, III, IV, V, and VI, the ATPase subunits, VI and VIII and cytochrome b (12,69,71). Finally, it should be noted that increased wobble in the 5' position of anticodons in mitochondrial tRNAs exists. This has resulted in the partial deviation of mtDNA from the universal genetic code (72,73). DNA REPLICATION Replication of duplex helical DNA involves a complex, highly coordinated series of reactions in which new DNA chains are initiated and elongated on each strand of the duplex. The majority of the information regarding the molecular and biochemical requirements of DNA replication has been obtained from extensive studies of Escherichia coli and its viruses, as well as eukaryotic viruses (See 74,75 for review). Intensive genetic and biochemical analyses have indicated that much of the host cell DNA replication machinery is required for viral DNA replication, thereby allowing for the identification and characterization of these enzymes. DNA replication is a semi- conservative process that usually initiates at a specific DNA 19 sequence (76). This process is regulated at the point of initiation, may be unidirectional or bidirectional and proceeds in a 5'-3' direction. PRORARYOTIC DNA REPLICATION The biochemical studies of DNA replication in prokaryotic systems have defined several proteins required for the initiation of DNA synthesis and subsequent elongation of the DNA chain. This progress came largely from the cloning of the unique 245 base pair chromosomal origin from E. coli (oriC) in plasmids and the isolation and analysis of the multiprotein system that replicates these plasmids (77). These proteins fall into three groups: (i) initiation proteins (dnaA, dnaB, dnaC, HU protein, and RNA polymerase) that recognize supercoiled oriC , alter its structural conformation, and lead to its further opening by dnaB helicase activity; (ii) specificity proteins (topoisomerase I, ribonuclease H and protein HU) that suppress potential replication origins elsewhere on the chromosome; (iii) replication proteins (DNA primase, DNA polymerase, gyrase, helicase, single-stranded binding protein and DNA ligase) that prime and elongate chains at the opened replication fork. In E. coli, the product of the dnaA gene is required for the initiation of DNA replication in vivo (78) and in vitro (79). DnaA protein has been shown to specifically and cooperatively bind four recognition boxes that are nine nucleotides in length with the consensus sequence TTATA/CCAA/CC (80). Upon binding, dnaA appears to induce a localized 2() unwinding of oriC in an ATP-dependent manner to create a structure required for subsequent steps of replication (81). The subsequent binding of a dnaBg-dnaCG-ATPg complex in the open DNA complex allows for continued opening of the DNA (82). The DNA-dependent hydrolysis of ATP results in the release of dnaC, allowing dnaB to function as a helicase on the DNA (83). Following sufficient opening of the replication fork, RNA polymerase or the primosome (dnaG, n, n',n", and i) generates a primer of sufficient length for utilization by the DNA polymerase to replicate the DNA molecule. Advancing of the replication fork in prokaryotes can be depicted in two models (for review see 84, Figs. 4 and 5). Both depend upon the opening of the DNA duplex by a helicase and the relief of consequent positive supercoiling by a topoisomerase or DNA gyrase. Single-stranded DNA binding protein protects the exposed DNA strands from nuclease degradation and from reannealing. In the first model (Fig. 4), replication proceeds by continuous synthesis on the leading strand and discontinuous synthesis on the lagging strand (76). The second model (Fig. 5) depicts concurrent replication of the leading and lagging strands and hypothesizés the priming of the nascent fragments of the lagging strand integrated with continuous synthesis of the leading strand. This scheme for concurrent replication requires looping of the lagging strand template by 180°, endowing to this strand the same orientation as the leading stand at the replication fork. A primer for the lagging strand is generated 21 Figure 4 Scheme for enzymes operating at a replication fork of E. coli. Proposed in Kornberg, A. (1982) DNA Replication (suppl.) W.H. Freeman and Co., New York p. 122. 22 z z PRIMOSOME 5 L__13 O; . 0 rep PROTEIN b--.- - 3:3 (HELICASE) j'nl n”. "if :ydnaa dnac complex DNA BINDING /;l PRIMASE PROTEIN (SSB) POLYMERASE Ill HOLOENZYME DNA \—”® ___/ POLYMERASE I LIGASE --—"l ‘3: 31% LEADING LA GGING STRAND STRAND Figure 4 23 Figure 5. Scheme for concurrent replication of leading and lagging strands by an asymmetric and dimeric DNA polymerase associated with a primosome and a helicase. Proposed in Kornberg, A. (1982) DNA Replication (suppl.) W.H. Freeman and Co., New York p. 125. 24 HELICASE POLYMERASE lll PRIMOSOME / HOLOENZYME PRIMERS 3' j _—. 5' Figure 5 25 by a DNA primase and extended by the DNA polymerase as the lagging strand is drawn past the primer. Evidence for this model includes the identification of the asymmetric dimer form of E. coli DNA polymerase III holoenzyme at the replication fork (85), the complexing of primase by some polymerases (86) and the replication of both stands of duplex DNA by phage T4 DNA polymerase with kinetics consistent with the action of one DNA polymerase molecule (87). EUNARYOTIC DNA REPLI CAT ION Much of the impetus for studying DNA replication in animal virus genomes comes from a desire to understand the events that occur during the replication of eukaryotic chromosomes. Simian virus 40 (SV40) has proven to be an excellent model system for studying the mechanisms of cellular DNA replication (75,88). SV40 DNA replication takes place in the nucleus of the host cell where the viral genome is complexed with histones to form a nucleoprotein structure (minichromosome) indistinguishable from cellular chromatin. Since SV40 encodes only a single replication protein (T antigen), the virus makes extensive use of cellular replication machinery. Thus, many similarities between viral and cellular DNA replication exist. In both cases, initiation of DNA synthesis results in the establishment of two replication forks that move in opposite directions. At each fork one of the two nascent strands (the leading strand) grows continuously. The other strand (the lagging strand) grows discontinuously by joining together small (ca. 200 bp) segments 26 of DNA that are independently initiated with RNA primers. Completion of DNA replication occurs when the two opposing forks meet. The development of a cell-free DNA replication system by Li and Kelly (89) has allowed for the identification and functional characterization of components of the cellular replication apparatus. Experimentation with this in vitro system allowed the development of a model describing the eukaryotic DNA replication fork (Fig. 6; see 88 for review). The initiation events of eukaryotic chromosomal DNA replication cannot be defined, primarily due to the inability to isolate eukaryotic chromosomal origins. However, study of SV40 DNA replication has allowed for the identification of several proteins involved in the initiation of eukaryotic DNA replication, as well as their characterization. Initiation of SV40 DNA replication requires a number of ATP-dependent reactions at the SV40 origin of replication carried out by T- antigen (90). This leads to DNA unwinding in the presence of single-stranded DNA binding protein (SSB) and topoisomerase I. T-antigen appears to bind DNA in two discrete reaction steps requiring 12 T-antigen monomers. This binding has been shown to be greatly facilitated by the addition of ATP(91,92) resulting in the disruption of the local structure of the DNA (93). Following the formation of the T-antigen-origin complex, T- antigen, in the presence of ATP and replication factor A (RF-A), catalyzes the unwinding of the duplex DNA in the 3'-5' direction 27 Figure 6. Model of the SV40 DNA replication fork and the enzymes involved. TAg, T-antigen; Topo I, topoisomerase I; DNA Pol a-primase complex, DNA polymerase d-primase positioned on the lagging strand; HSSB, human single-stranded binding protein; P01 6, DNA polymeraselb positioned on the leading strand; PCNA, proliferating cell nuclear antigen; ATP, adenosine S'triphosphate. Taken from Hurwitz, J., Dean, F.B., Kwong, A.D. and Lee, S.H. (1990) J. Biol. Chem. 265, p.18044. I Repllcetlon Fork W 7‘9 (3'-5') DNA pol8 complex (poi 8. scum. me: I. «see. am 28 “2 0'! Topo I /(reIIeve posItIve superheIIcIty) .IV'IIIAI"'III.AII'IIMIMII..II"'II..IV"I... DNA polo -prImase complex Figure 6 29 (93,94). This unwinding is stimulated by the dephosphorylation of T antigen by protein phosphatase 2A (95). A major role of RF-A is the stabilization of the exposed single strands (96). Following the initial unwinding of the duplex at the origin region, short DNA chains are synthesized by the action of the DNA polymerase a-primase. This enzyme plays a critical role in DNA replication in the in vitro system because it appears to contain the only activity capable of starting the nucleotide chains de novo (97). SV40 DNA replication is completely dependent upon the presence of the a polymerase-primase. In the absence of Pol a, DNA synthesis in the SV40 system is not observed. In the absence of Pol 5,DNA products less than 500 bp in length are observed, yet the SV40 genome is entirely replicated. DNA polymerase a-primase synthesizes short RNA- primed DNA fragments on the lagging strand in the presence of human SSB. Joining of these "Okazaki fragments" to form a completed daughter strand requires the removal of the RNA primers by a 5'-3' exonuclease and RNase H, filling of the generated gaps by a DNA polymerase and sealing of the nicks by DNA ligase (98). In the presence of both DNA polymerases and the entire complement of purified enzymes required for SV40 DNA replication (T antigen, RF-A, PPZAC {the catalytic subunit of protein phosphotase 2A) PCNA, replication factor C and topoisomerase I and II) complete replication of the SV40 genome <3ccurs with observable leading (continuous) and lagging (discontinuous) DNA synthesis (97). Finally, topoisomerase II DNA sequence in this species appears to be precise and 35 complete. The second species is entirely RNA, with the same 5'terminal sequence as the first species and with the same 5'-3' orientation towards the tRNAPr° gene. This suggests that both priming of H-strand replication and transcription of L-strand genes begin at the same sequence in the genome. Further, the transition to DNA synthesis appears to be a later event. The LSP may therefore serve a dual function: the generation of RNA primers for DNA synthesis at the OH and the initiation of transcription for the L-strand genes in the mitochondrial genome. The mechanism(s) modulating the RNA-DNA transition, that is, whether or not the RNA molecule originating from the LSP becomes a full length transcript or a primer for DNA synthesis, has not been fully elucidated. However, a mitochondrial RNA-processing endoribonuclease (RNase MRP), together with a nuclear encoded 135 nucleotide RNA moiety required for activity, has been characterized from mouse and human cells (109,110). RNase MRP cleaves mitochondrial RNA transcripts complementary to the L— strand of the displacement loop at a unique site between conéérved sequence blocks II and III. Further, the position of the mitochondrial substrate cleavage site utilized by the RNase MRP is conserved relative to the displacement loop sequence element (CSB II). Also, the RNase MRP RNA moiety bears a region complimentary to the CSB II sequence in the mtRNA substrate (106,111). Replication of Drosophila mtDNA is not characterized as well as that for mammalian mtDNA. Electron microscopic studies 36 indicate that replication of Drosophila mtDNA molecules also proceeds in a unidirectional manner (Fig. 8, 63). The process is more asymmetric than in mammalian systems as the leading strand is 87-95% complete prior to the initiation of the lagging strand (54). The origin of replication of both DNA strands resides within the non-coding A+T rich region, with replication of the leading strand proceeding in the direction of the genes encoding the two ribosomal RNAs (54,61,63,112,113). The A+T rich regions of D. yakuba and D. virilis mtDNA have been sequenced and compared (113). D. yakuba and D. virilis belong to the subgenera sopophora and drosophila, making them distantly related (114). The position of the putative origin of replication in these two species is conserved even though the surrounding DNA sequence is divergent. In addition, the origin region of both Drosophila species has the potential to form a stable hairpin loop structure similar to the mammalian OL (51,76,113). This structure may be used as the site of initiation of lagging strand synthesis. PRONARYOTIC DNA. POLYMERASES All DNA polymerases isolated to date, whether from bacterial, viral, or eukaryotic sources, have the following properties: (1) chain growth in the 5'-3' direction, anti-parallel to the template strand; (ii) template-directed base selection via formation of the appropriate Watson-Crick base pair with the incoming deoxyribonucleotide; (iii) inability to initiate DNA chains de novo. 37 Figure 8. A scheme for the replication of Drosophila mitochondrial DNA. Thick and thin continuous lines represent the complementary parental strands. Thick and thin broken lines represent the corresponding complementary daughter strands. The arrows on the daughter strands show the direction of DNA synthesis (assuming antiparallel synthesis of the two complementary strands). In a, the origin of replication, (0) and direction of replication (R) around the molecule are indicated. Modified from Wolstenholme, D.R., Goddard, J.M. and Fauron, C.M.-R. (1979) In: Proc. 8th Ann. I.C.N;-U.C.LMA. Symp. on Mblecular and Cellular Biology: Extrachromosomal DNA, Vol.1 (eds. D. Cummings, P. Borst, I.B. Dawid., S. weisman, and C.F. Fox), Academic Press, New York p. 420. 38 DROSOPHILA MITOCHONDRIAL DNA REPLICATION Figure 8 39 Escherichia coli DNA is predominantly maintained by two DNA polymerases: DNA polymerase I (Pol I) and DNA polymerase III holoenzyme (Pol III HE) (76). The former is a multifunctional enzyme involved in both DNA repair and replication. The latter is a complex replicative enzyme responsible for the bulk of the replication of the E. coli chromosome. Both enzymes absolutely require a divalent cation (usually Mgz+) for DNA polymerase activity and have a low salt optimum (20 mM KCl). DNA polymerase I has three enzymatic activities: a 5'-3' DNA polymerase, a 3'-5' exonuclease and a 5'-3' exonuclease. Pol I is a relatively nonprocessive enzyme, incorporating approximately 10 nucleotides per binding event (76). In terms of accuracy of deoxynucleotide selection, Pol I has an in vitro error rate of one in 105 base pairs replicated (115). It has a molecular weight of 109 kDa and its complete amino acid sequence is known (116). DNA polymerase III HE is a complex of seven distinct subunits (a, t, ‘y, B, 5, e, 9, see references 117 and 118 for a review), each required for full reconstitution of holoenzyme activity in vitro. Further, Pol III HE has been proposed to function as an asymmetric dimer (85), suggesting differential function for leading and lagging strand DNA synthesis. DNA POlymerase III HE is a highly processive and accurate DNA Polymerase. Under optimal conditions, the holoenzyme form of PC11 III will incorporate greater than 5000 nucleotides per tfiinmdate association-catalysis-dissociation event (119). The fiudelity of DNA synthesis by Pol III HE in vitro has been shown 40 to approach the in vivo error rate of E. coli DNA synthesis of 1 error in 107 nucleotides polymerized (120). Some of the subunits comprising Pol III HE have been assigned specific functions. The a-subunit is a 130,000 dalton polypeptide, contains the 5'-3' DNA polymerase catalytic activity and is encoded by the dna E gene (121). The B-subunit is a 37 kDa polypeptide and is the product of the dna N gene (122). This subunit is required for initiation complex formation, increases the processivity of the core enzyme complex from 200 nucleotides to >5,000 nucleotides (123) and binds Mg2+ (124). The 8-subunit is encoded by the dna Q gene, contains the 3'-5' proofreading exonuclease activity and has a molecular mass of 27,500 daltons (125). Neither the functions nor the structural genes for the 9- and 5-subunits have been identified. The t- and.Ybsubunits are both encoded by the dna X gene, where the ybsubunit is produced by a -1 ribosomal frameshift of this open reading frame (126). The molecular weights of t and y are 71,000 and 52,000 daltons, respectively. Deletion analysis and partial tryptic digests indicate that t and.1 share common amino acid sequences and that 7 is completely included within the amino-terminal two thirds of r (127,128). In terms of functional properties, the t-and.ybsubunits have been shown to bind the ATP/dATP required for initiation complex formation (129) and are essential for the high processivity of Pol III HE (130). The t-subunit is a DNA-dependent ATPase and presumably binds DNA in its C-terminal domain which is very basic. This putative ligand binding site is absent in the y-subunit (129). 41 Further, the t-subunit is responsible for the dimerization of DNA polymerase III (131). Unlike simpler DNA polymerases, the Pol III HE catalyzed reaction has several distinct reaction states. In the presence of ATP or dATP a tight stable initiation complex forms, accompanied by ATP hydrolysis to ADP presumably by the t— and 7- subunits (129). This ATP binding is enhanced by the B-subunit (132). The addition of the four dNTPs to the initiation complex in the presence of Mg2+ results in rapid and processive DNA synthesis (123). The mechanism of displacement of the enzyme from the DNA has not been determined. However, in vitro termination complexes, (resulting from complete replication of a circular template) are stable for more than 40 minutes in the presence or absence of ATP (133). EUNARYOTIC DNA. POLYMERASES DNA polymerase a (Pol a) is currently thought to be primarily responsible for the replication of the lagging strand at the DNA replication fork (88). The a polymerase is distinguished from other cellular DNA polymerases by its sensitivity to aphidicolin (163), butylanalinouracil (135) and N-ethylmaleimide, and by its resistance to inhibition by dideoxynucleoside triphosphates (136). Pol a displays optimal activity at a pH range of 7.5 to 9.0 and at a low salt concentration of approximately 20 mM KCl (see 137 for review). Further, Pol a has a relatively high affinity for dNTPs (Km, 4-20 uM). The enzyme, as purified from several sources, is a tetramer composed of four non-identical 42 subunits. The 182 kDa subunit contains the 5'-3' DNA polymerase activity and, in the Drosophila enzyme, a cryptic 3'-5' mispair specific exonuclease that is unmasked following removal of the 73 kDa polypeptide (86,138). The 50 and 60 kDa subunits together provide for a DNA primase activity. The 73 kDa subunit has not been assigned a function (87,137). The affinity of Pol a for any given template is variable and dependent upon the presence or absence of specific competing DNAs (139). This suggests that the introduction of a second polynucleotide influences binding of the first polynucleotide allosterically (140). Further, enzymological studies of Pol a have revealed the presence of two DNA binding sites and an ordered sequential mechanism of substrate binding and product release (140,141). This mechanistic scheme prescribes that Pol a.binds the template as the first substrate, followed by the primer as the second substrate and the dNTP as the third. The a polymerase may then translocate along the template, incorporating 8-20 nucleotides prior to dissociation. The fidelity of DNA replication in vitro by the Drosophila a polymerase was determined as 1 error per 118,000 nucleotides incorporated (142). The fidelity of its isolated 182 kDa a- subunit (polymerase subunit) is 80- to ZOO-fold more accurate than the intact polymerase. This indicates that the 3'-5' exonuclease functions as a proofreading enzyme (138). The intrinsic DNA primase activity of Drosophila Pol a has been separated from the DNA polymerase by treatment with urea or ethylene glycol, followed by glycerol gradient sedimentation in 43 the presence of the same chaotropic agents (86,138). This enzyme activity shows a Mg2+ optimum of 2 to 6 mM (143), is salt sensitive (activity is 30% of optimal in the presence 15 mM NaCl) and exhibits a relatively high Km for nucleoside triphosphates (3 to 5 mM). When it is a part of the Pol a holoenzyme, the primase synthesizes a primer of 12-14 nucleotides inulength. However, when dissociated from the 182 and 73 kDa subunits, the primase synthesizes a product 24-28 nucleotides long. In the SV40 DNA replication system, the mouse hybridoma primase has been shown to initiate primers at one major site for each strand (144). The in vivo expression of Pol a is characterized by ~increased levels of the enzyme during the G1 and S-phases of the cell cycle, in which the cell prepares to and replicates its DNA (145). However, the presence of nuclear DNA polymerase a is not restricted to S-phase. Therefore, the potential exists that Pol a.may exist as a phosphoprotein. In a study using Pol a from rat embyronic fibroblasts, a 220 kDa polypeptide was identified as a phosphoprotein. Here, the polymerase activity was reduced greater than 10-fold upon dephosphorylation (146). In addition, enzymatic activity could be restored upon incubation with ATP. DNA polymerase 5 is currently thought to be the enzyme responsible for synthesis of the leading strand at the DNA replication fork (97). This enzyme is characterized by its sensitivity to aphidicolin and inhibition by N-ethylmaleimide but resistance to dideoxynucleoside triphosphates. While the drug butylphenyl dGTP strongly inhibits the a polymerase, the 5 44- polymerase is inhibited only marginally (148,149). In addition, studies with monoclonal antibodies developed against Pol a have been shown not to cross-react with Pol 5 (150). Pol 5 is functionally distinguished from the a polymerase by its association with a proofreading 3'-5' exonuclease activity regardless of the tissue source (151,168,169). The 5 polymerase is a highly processive DNA polymerase which, in the presence,of PCNA and RF-C, will incorporate several thousand nucleotides per binding event (101). In addition, Pol 5 is a highly accurate DNA polymerase, producing on average <1 error per 106 nucleotides polymerized, an error rate 10-fold lower than that of a polymerase (151). The subunit structure of the 5 polymerase has not yet been clearly defined but recent evidence indicates that the enzyme is a heterodimer comprising subunits of 130 and 50 kDa (see reference 152 for review). Most recently a fifth eukaryotic DNA polymerase has been studied and termed Pol 8 (see references 152 and 153 for review). Its subunit structure has not yet been determined. However, the catalytic subunit is proposed to be 225 kDa. Functionally, the e polymerase is characterized as a highly processive, PCNA-independent DNA polymerase with an associated 3'-5' exonuclease activity (153, 154). The role Pol t plays at the replication fork has yet to be established. However, a conditional-lethal mutant of budding yeast DNA polymerase s has been isolated, suggesting Pol 6 plays an essential role in DNA replication (153). 45 DNA polymerase B has a synthetic role in DNA repair. In terms of size and catalytic repertoire, Pol B is the simplest DNA polymerase. The human and rodent enzymes are monomers with molecular weights of 39 kDa (155). The DNA polymerase activity is fully distributive and is the only activity associated with this enzyme. The base substitution error frequency of the HeLa cell B polymerase is greater than one mistake for every 5000 nucleotides polymerized (156), making it the least accurate of the higher eukaryotic DNA polymerases. The B polymerase follows ordered bi-bi kinetics in a two substrate, two product reaction (155). Distinguishing Pol B from other DNA polymerases is its resistance to the sulfhydryl blocking agent N-ethylmaleimide (76). The human gene encoding the B polymerase has been cloned, and shown to be present in a single copy on the short arm of chromosome 8 (157). The activity of DNA polymerase B may be post-translationally regulated. An in vitro study demonstrated that protein kinase C phosphorylates two amino acid reSidues, Ser44 and Ser55 (147). It is not known if these amino acid residues are located within ligand binding domains. Phosphorylation by protein kinase C results in the inactivation of Pol B activity. Further, the recovery of Pol B DNA polymerase activity was achieved by dephosphorylation with alkaline phosphatase. DNA polymerase 7 is a nuclear-encoded DNA polymerase and is the sole DNA polymerase found in mitochondria (39). Thus, it is essential for mtDNA replication and mitochondrial biogenesis. Pol 7'has been highly purified from four sources: chick embryos 46 (158), Drosophila embryos, (159), xenopus laevis oocytes (160) and porcine liver (161). The 7 polymerase, as purified from each source,is inhibited N-ethylmaleimide and by dideoxynucleoside triphophates, distinguishing it from Pol a, 5, e and B. The 1 polymerase is also distinguished from the other cellular DNA polymerases by its relatively high monovalent cation optimum of 200 mM on activated calf thymus DNA (159,161) and >100 mM on poly dA - oligo dT (158-161). In addition, a 3'- 5' exonuclease has been shown to be associated with the Y polymerase as purified from each of the above sources (161-164). The mtDNA polymerase has been shown to be a highly accurate DNA polymerase (161,162,167). This is necessary for the faithful DNA replication of the 13 genes required for oxidative phosphorylation and electron transport which are encoded by mtDNA (12,69,70). However, a paradox exists. The rate of evolution of animal mtDNA is 5- to lO-fold greater than that of single copy nuclear DNA (165), yet the mtDNA polymerase replicates DNA accurately. The apparent absence of DNA repair in mitochondria (166) suggests that accurate replication of the mitochondrial genome may be solely dependent upon its replication machinery.) The accuracy of DNA polymerization by the Drosophila y polymerase is nearly identical to that of Drosophila Pol a and E. coli DNA polymerase III HE in the presence of SSB (167). However, this was determined in an assay which is not capable of detecting all possible mispairs. DNA polymerase m purified from porcine liver and chick embryos, has been examinined in such an 47 assay (161,162,168) and error frequencies of 1 per 500,000 and 1 per 260,000 nucleotides incorporated were demonstrated, respectively. The Drosophila mtDNA polymerase is a quasi-processive enzyme, incorporating 25 to 45 nucleotides per binding event on singly- primed ¢X174 DNA in the presence of 110 mM KCl (167). In addition, Drosophila Pol 7 appears to preferentially utilize previously utilized template-primer termini. Notably, in the absence of KCl, the T'polymerase will replicate the entire M13 genome (10,130 nt) in a single binding event (Wernette and Kaguni, manuscript in preparation). The complete replication of the M13 template-primer has been accomplished by the near- hombgeneous Fr.VI enzyme and by the 20% pure Fr.IV enzyme. This suggests, unlike the situation for Pol 5 and E. coli Pol III HE, that high processivity may be an intrinsic property of the y polymeraSe. In this regard, the near-homogeneous chick embryo Y'polymerase has been shown to incorporate up to 4000 nucleotides per binding event using the synthetic homopolymer poly rA - oligo dT (158). In terms of template-primer utilization, Drosophila y polymerase is a versatile enzyme that replicates both double- stranded and single-stranded DNA, whereas Pol 0 cannot efficiently utilize a predominantly single-stranded DNA template (167). The ability of Pol y‘to efficiently utilize Predominantly single-stranded DNA in vitro, either by the highly Processive or quasi-processive mechanism, is consistent with the 48 replication of the mitochondrial genome in asymmetric and continuous manner. The physical structure of the y’polymerase has not been firmly established. The chick embryo y'polymerase has a sedimentation coefficient of 7.5 and elutes from Sephadex G-200 consistent with a molecular weight of 180 kDa (158). SDS- polyacrylamide gel electrophoresis of the highly purified preparation demonstrated the presence of two polypeptides of 47 kDa (major polypeptide) and 135 kDa (minor polypeptide), both proportional and coincident with DNA polymerase activity. The near-homogeneous Drosophila y'polymerase has a native molecular weight of approximately 160 kDa as determined by velocity sedimentation and Sephacryl S-200 gel filtration chromatography. SDS-polyacrylamide gel electrophoresis indicated this enzyme is a heterodimer composed of a 125 kDa a-subunit and a 35 kDa B- subunit (159). Analysis of DNA polymerase activity in situ demonstrated that the 125 kDa subunit contains the DNA polymerase activity. To date, the porcine liver y'polymerase has not been purified to near-homogeneity. However, glycerol gradient sedimentation provides an S-value consistent with a globular enzyme with a native molecular mass of 160 kDa (161). Subsequent SDS-polyacrylamide gel electrophoresis demonstrated the presence of polypeptides with molecular weights of 120, 55, 50 and 48 kDa. The xenopus 7 polymerase has a native molecular mass of approximately 180 kDa as determined by sedimentation analysis (160). SDS-polyacrylamide gel electrophoresis indicates the presence of six abundant polypeptides with 49 molecular weights of 140, 100, 85, 55, 40, and 31 kDa. Cross- linking to a radiolabeled and BrdUMP substituted oligonucleotide, presumably to the template-primer binding domain, resulted in the labeling of the 140, 100, and 55 kDa polypeptides. Taken together, the data suggest that DNA polymerase 7 is a heterodimer with a 125-140 kDa catalytic subunit and a 31-47 kDa subunit. 3'-5' PROOFREADING EXONUCLEASES One of several mechanisms that ensures maintenance of the integrity of genetic information is proofreading during DNA replication. This is the 3'-5' exonucleolytic reversal of polymerization to remove nucleotides that have been incorrectly inserted during DNA synthesis. A proofreading exonuclease generally exhibits the following characteristics: it prefers single-stranded to double-stranded DNA substrates; it preferentially excises a mispaired rather than correctly paired primer terminus; it is physically associated with the DNA polymerase, either as a part of the same polypeptide or as an associated subunit; and it acts coordinately with the polymerase to enhance the fidelity of DNA synthesis. The contribution of a 3'-5' proofreading exonuclease to the fidelity of DNA polymerization can be estimated by comparing the accuracy of DNA synthesis under conditions that either allow or inhibit 3'-5' exonuclease activity (for review see 169). Such approaches have demonstrated that proofreading increases the fidelity of DNA polymerase I about 10-fold (170). Site-directed 50 mutagenesis studies of E. coli DNA polymerase I Klenow fragment, resulting in Exo‘ derivatives of the enzyme, demonstrated that the exonuclease improves the average base substitution fidelity by 4- to 7-fold (115). This indicates the primary mechanism of faithful DNA synthesis by E. coli Pol I is accurate nucleotide selection. The contribution of the 3'-5' proofreading exonuclease to the fidelity of E. coli DNA polymerase III HE has been estimated to be approximately ZOO-fold. (120). The functional coordination of the e-subunit, which contains the 3'-5’ exonuclease, with the a-subunit, the DNA polymerase subunit, is demonstrated by a 26- fold increase in affinity for 3'-OH termini and a 10- to 80-fold increase in exonuclease activity upon association of the subunits (171). The DNA polymerase of bacteriophage T4 contains the DNA polymerase and 3'-5' exonuclease activities on the same polypeptide. The error frequency of this DNA polymerase approaches 1 per 107 nucleotides incorporated (172). The 3'-5' exonuclease is highly active on mispaired substrates (173), and can be inactivated by selective oxidation (174). Following such inactivation, the fidelity of the enzyme was shown to decrease approximately 9-fold. Higher eukaryotes contain five DNA polymerases (a,(L n.8,e) as previously mentioned. The 3'-5' proofreading exonucleases associated with Pol.0t as isolated from Drosophila embryos and HeLa cells, and Pol % as isolated from Drosophila embryos and porcine liver, have been shown to be highly mispair-specific 51 (138,161,163,l74). The proofreading exonucleases associated with the chick embryo and the xenopus oocyte mtDNA polymerases exhibit preferential excision of mispairs, but are not highly mispair-specific (162,164). The contribution of the proofreading 3'-5' exonucleases associated with eukaryotic DNA polymerases to the fidelity of DNA synthesis has been clearly demonstrated with the Drosephila a polymerase (138). Here, the 3'-5' exonuclease increases the fidelity of DNA synthesis two orders of magnitude. In the case of DNA polymerase 5 purified from calf thymus, proofreading by the associated 3'-5' exonuclease appears to improve the fidelity of DNA synthesis as much as 100-fold (151). A more complete understanding of proofreading requires a kinetic description of the editing reaction and factors that influence it. For example, the nucleotide composition of the mispair and the surrounding DNA sequence may exert differential effects which influence the balance between the competing processes of polymerization and excision. The polymerase inserts a base at a rate characteristic of the base pair or mispair formed, and then either the polymerase incorporates the next correct nucleotide, or the exonuclease excises the terminal base prior to further polymerization. The overall fidelity of the reaction is dependent on the rate of misincorporation, the rate excision of the mispair and the rate of extension of the mispair formed. To define the kinetics of proofreading, the relative reaction rates for formation and subsequent hydrolysis or extension of a mispair must be determined. 52 Another consideration is the mechanism of strand transfer between the separate DNA polymerase and 3'-5' exonuclease active sites. A study with E. coli DNA polymerase I Klenow fragment demonstrated that DNA dissociation is more rapid than exonucleolytic digestion (175). This indicates Pol I Klenow edits its own polymerase errors by a predominantly intermolecular process involving dissociation of the enzyme-DNA complex and reassociation of the DNA at the exonuclease site of a second Klenow molecule. Conversely, the T5 DNA polymerase appears capable of switching its direction from degradation in the 3'-5' direction to synthesis in the 5'-3' direction without leaving the primer-template (176). STRUCTURE-FUNCTION STUDIES OF DNA POLYMERASES [coli DNA POLYMERASE I A combination of structural, biochemical and genetic studies has led to the conclusion that Pol I comprises three domains. Limited proteolysis of the enzyme removes the 35 kDa N-terminal domain that contains the 5'-3' exonuclease (see reference 76 for review). The remaining 68 kDa polypeptide (Klenow Fragment) contains the 3'-5' exonuclease and DNA polymerase activities. X-ray crystallographic studies have shown that the Klenow fragment is folded into two domains (177). The DNA polymerase active site is located on the larger C-terminal domain (residues 521-928) that is in the form of a deep cleft about 2.4 nm wide and 3 nm deep. Approximately 3 nm from the DNA polymerase 53 domain is the 3'-5' exonuclease active site on the N—terminal domain (residues 324-517). Photoaffinity crosslinking of the Pol I Klenow dNTP binding site using 8-azido ATP (178) and Schiff base crosslinking of pyridoxal phosphate (which binds competitively to the dNTP binding site) (179) resulted in the labeling of Tyr766 and Lys758, respectively. Both amino acid residues are located on helix O of the Klenow fragment, indicating its importance with regard to dNTP binding. Site-directed mutagenesis studies revealed that the amino acid residues Tyr765, Arg341, and Asn845 appear to contact the incoming dNTP, and that residues Gln849 and Arg668 appear to play a role in the catalysis of the polymerase reaction (180). Notably, these amino acids are positioned deep within the proposed DNA binding cleft. Further, the amino acid residues Asp355, Glu357 and Asp424 have been shown to be critical for 3'— 5' exonuclease activity (181). E.coli DNA POLYMERASE III Analysis of structure-function relationships in Pol III HE has been limited to isolating active subassemblies of the subunits and to obtaining genetic evidence for gene product interaction. These studies have revealed several subunit- subunit interactions. The proposed subunit stoichometry is that of an asymmetric dimer (a2, 91.71:: B2,, 52, £2, 82) . The a- subunit interacts with e,(171) B, (182,183), 8 (171) and itself in the absence of other proteins (118). The B-subunit is known to interact with itself, since it can be isolated as a dimer 54- (184). Further, the B-subunit has been shown to interact with t and Y (185). The 5-subunit has never been purified in a form free of 7, indicating a close association of these two polypeptides. Further, the 7-5 complex can functionally interact with DNA polymerase III (117). The t-subunit may interact with 5, but no direct evidence is available. The 1- subunit does associate with one of the core DNA polymerase III components (131). The exact functional relationship of t to Y is not yet clear. Both are derived from the dna X gene and lead to the formation of the asymmetric dimer. The presence of y'and t within the holoenzyme could directly impose different functional properties upon each half. For example, 1, when added to DNA polymerase III in the absence of the other subunits, increases the processivity 6-fold (123). In addition, the t-subunit binds single-stranded DNA (129). This would argue for the association of t with the leading strand half of the holoenzyme. On the other hand, 1&5 in association with ATP, the B-subunit and other subunits excluding‘n allow for highly processive DNA synthesis (186), possibly suggesting that the t-subunit is associated with the lagging strand half of the DNA polymerase. STRUCTURAL STUDIES IN EUNARYOTIC DNA POLYMERASES The amount of information regarding structure-function relationships in eukaryotic DNA polymerases is limited. Subunit separation studies of Drosophila a polymerase, using ethylene glycol (138) and urea (86) demonstrated that the 182 kDa subunit 55 contains the DNA polymerase and the cryptic 3'-5' exonuclease activities; the 50 and 60 kDa subunits possess the primase activity. The predicted amino-acid sequence derived from a cDNA clone of the 180 kDa subunit of the human a polymerase has allowed preliminary localization of the deoxynucleoside triphosphate and DNA binding domains following comparison with the amino acid sequences of both viral and bacteriophage DNA polymerases (187). Of the eukaryotic DNA polymerases, only Pol B has been cloned and the recombinant enzyme expressed in an active form (152). The amino acid and nucleotide sequence suggest a globular enzyme composed of seven a-helical regions and three regions of B-sheet (188). Site-directed mutagenesis studies examining the active site domains have not been reported. I An in situ DNA polymerase experiment with the Drosophila y polymerase demonstrated that the 125 kDa subunit is the DNA polymerase catalytic subunit (159). Subunit association of the 3'-5' exonuclease remains undefined as is the function of the B- subunit. DNA binding site analysis of the xenopus oocyte Pol 7 indicated the 140 kDa subunit is most likely the catalytic subunit (160). Glycerol-gradient sedimentation of the frog enzyme in the presence of 50% ethylene glycol led to the conclusion that the 140 kDa polypeptide contained both the DNA polymerase and 3'-5' exonuclease activities (164). However, the Presence of polypeptides of approximately 40 kDa in molecular weight in the 140 kDa polypeptide containing fractions indicates 56 a lack of subunit separation, thereby invalidating the latter assignment. CHAPTER II EXPERIMENTAL PROCEDURES MATERIALS Nacleotides and nucleic acids. Unlabeled deoxy- and ribonucleoside triphosphates were purchased from P-L Biochemicals. Unlabeled 5'-nucleoside monophosphates were purchased from Sigma. [3H]dTTP, [3H]dCTP and [3H]dATP were purchased from ICN Biochemicals; [a-32P1dCTP and [y-32P1ATP were purchased from New England Nuclear. Nucleoside monophosphates were purchased from Sigma. Calf thymus DNA (highly polymerized Type I) was purchased from Sigma and was activated by partial digestion with DNase I (Boehringer Mannheim) as described by Fansler and Loeb (189). Poly (dA)7oo—p(dT)1o was purchased from P-L Biochemicals and contains adenine and thymine in a molar ratio of 20:1, respectively, such that the average single-stranded DNA region between primers is 200 nucleotides (nt). A recombinant M13 viral DNA (10,650 nt) was prepared by standard laboratory methods. (dT)15 and synthetic oligodeoxynucleotides (15 nt or 17 nt) complementary to M13 viral DNA and containing a 3'-terminal basepair (dAMPtemplate:dTMPprimer or dGMP:dCMP)or a 3'-terminal mispair (dAMPszMP, dGMP:dAMP or dGMP:dGMP) were synthesized in an Applied Biosystems Model 477 oligonucleotide synthesizer. Poly(dA)-oligo(dT)uy{3H](dC)1, - (dA)1, and -(dT)1 were prepared by the extension of (dT)15 by terminal deoxynucleotidyl transferase in the presence of [3H]dCTP, dATP or dTTP, respectively (190). Poly(dAQ7oo'was subsequently 57 58 annealed to the 3'-terminally extended (dT)15 for 25 minutes at 553: using a 2:1 molar ratio of adenine to thymine in buffer H containing 10 mM Tris-HCl (pH 8.0), 0.3 M NaCl and 0.03 M sodium citrate. Enzymes and protein standards. Drosophila DNA polymerase y (Fractions I-VI) was prepared as described by Wernette and Kaguni (159). E. coli DNA polymerase I (Pol I) and its Klenow fragment, and T4 polynucleotide kinase were purchased from New England Biolabs. Terminal deoxynucleotidyl transferase was purchased from International Biotechnologies, Inc. E. coli DNA polymerase III holoenzyme (Pol III) (Fraction V, 50% pure) (117) was a gift from Dr. Jon Kaguni of this department. Bovine serum albumin, bovine carbonic anhydrase and prestained SDS molecular weight protein markers were purchased from Sigma. Goat anti-rabbit IgG (H+L) horseradish peroxidase conjugate was purchased from BioRad. Protein A-alkaline phosphatase conjugate and protein A-agarose were purchased from Sigma and Boehringer Mannheim, respectively. Chemicals. Phenylmethylsulfonyl fluoride (Sigma) was prepared as a 0.1 M stock solution in isopropyl alcohol and stored at -20°C. Sodium metabisulfite (J.T. Baker Chemical Co.) was prepared as a 1 M stock solution, adjusted to pH 7.5 with NaOH and stored at -20°C. Leupeptin was purchased from the Peptide Institute, Minoh-Shi, Japan, and was prepared as 1 mg/ml stock solution in 0.1 M potassium phosphate buffer, pH 7.5, and stored at -20°C. Dithiothreitol (DTT) was purchased from Sigma, prepared as a 1 M stock and stored at -20°C. Nitro blue tetrazolium and 5-bromo-4- chloro-3-indolyl phosphate were purchased from Sigma, prepared as .59 and 50 mg/ml stocks and stored at 4°C. Acetone (HPLC grade) was purchased from Mallinckrodt. Ethylene glycol was purchased from Fisher Scientific. Guanidine hydrochloride (Gu-HCl, ultra pure) was purchased from ICN. N-chlorosuccinimide was purchased from Sigma. METHODS DNA polymerase y'assay. Reaction mixtures (0.03 or 0.05 ml) contained 50 mM Tris-HCl (pH 8.5), 4 mM MgC12, 20 mM DTT, 110-200 mM KCl, 400 ug/ml bovine serum albumin, saturating levels of template-primer, 30 MM (each) dATP, dCTP, dGTP and [3H]dTTP (approximately 3000 cpm/pmol), and enzyme. The saturating concentrations for the DNA template-primers and the optimal KCl concentrations for Pol 7 were: 180 NM DNase I-activated calf thymus DNA at 200 mM KCl and 180 or 60 BM poly(dA)-oligo(dT)1o and 12 or 50 BM singly-primed M13 DNA at 120 mM KCl. Pol I and Pol III were assayed in the presence of 20 mM KCl. Incubation was at 30°C for 20 minutes. One unit of DNA polymerase activity is defined as that amount that catalyzes the incorporation of 1 nmol of deoxyribonucleoside triphosphate into an acid insoluble material in 60 minutes at 30°C. Deoxynucleotide turnover assay. The turnover assay measures the DNA-dependent conversion of a dNTP to the corresponding monophosphate. Reaction mixtures (0.04 ml) were as described for the DNA polymerase assay on DNase I-activated calf thymus DNA, except that [ar32P1dCTP was used (30 BM; 15,000 cpm/pmol for Pol y and 5,000 cpm/pmol for Pol I and Pol III). Pol I and Pol III were assayed in the presence of 20 mM KCl. Incubation was for 20 6() minutes at 30°C. Reactions were terminated by the addition of EDTA to 20 mM. To determine the amount of DNA synthesis, 5 pl aliquots were precipitated in duplicate in 0.5 ml of 10% trichloroacetic acid containing 0.4 M sodium pyrophosphate. The DNA was collected by filtration on Whatman GF/C glass fiber discs. The filters were washed with 10 ml of 1 M HCl containing 0.1 M sodium pyrophosphate. After drying, the filters were counted using a toluene-based scintillant. To determine the amount of deoxynucleoside monophosphate formed, 3 ul aliquots were applied in triplicate to a thin layer polyethyleneimine plate together with unlabeled deoxynucleoside mono, di, and triphosphate markers. The thin layer plate was developed with 1 M formic acid and 0.5 M LiCl. The ultraviolet absorbing spots corresponding to the deoxynucleoside mono, di, and triphosphates were cut out and counted as described above. 3'-5' exonuclease assay on synthetic DNA substrates. Reaction mixtures (0.03 ml) contained 50 mM Tris-HCl(pH 8.5), 4 mM MgC12, 20 mM DTT, 120 mM KCl, 400 ug/ml of bovine serum albumin, 46 BM poly(dA)-oligo(dT)15-[3HJdC1 (1115 cpm/pmol),-[3H]dA1 (456 cpm/pmol), or -[3H]dT1 (735 cpm/pmol) or 24.5 HM [3H] poly(dT)33 (880 cpm/pmol) and enzyme. Pol I was assayed in the presence of 20 mM KCl. Incubation was for 30 minutes at 30°C. Aliquots (0.012 ml) were spotted in duplicate onto DE-81 filter paper (1 x 1 cm, Whatman). The filters were washed twice for 10 minutes in 100 ml of 0.3 M ammonium formate (pH 7.6), followed by a 5 minute wash in 40 ml of 95% ethanol. After drying, the filters were counted in a scintillation counter in a toluene-based scintillant. 61 Preparation of 3'-[3H]-labeled DNase I—activated calf thymus DNA for exonuclease assay. DNase I-activated calf thymus DNA, as described under "Materials," was 3'-end labeled in a reaction mixture (1.5 ml) containing 3.75 mmol activated DNA, 140 mM K cacodylate (pH 7.2), 30 mM Tris base, 1 mM CoClz, 1 mM DTT, 6 nmol [3HJdTTP (45 Ci/mmol) and 90 units of terminal deoxynucleotidyl transferase. Incubation was for 10 hours at 37°C. The labeled DNA was purified from the unincorporated [3H]dTTP by gel filtration chromatography on a biogel A-5M (100 - 200 mesh) column. Preparation of 5'-[32P]-labeled substrates for product analysis after exonuclease assay. 3'-terminal mismatched oligonucleotides (15 nt or 17 nt) as described under "Materials" were 5'-end labeled. The kinase reaction (0.04 ml) contained 50 mM Tris-HCl (pH 7.5), 10 mM MgC12, 15 mM DTT, [y-3ZPJATP (0.5 DM, 3000 Ci/mmol), 85 pmol (as nt) of oligonucleotide and 20 units of T4 polynucleotide kinase. Incubation was for 60 minutes at 37°C. M13 viral DNA was added to a concentration of 6 mM (in 4-fold molar excess over homologous oligonucleotide). The DNA mixture was precipitated with ethanol. The DNA pellet was resuspended in buffer H (0.3 ml) and was incubated at 65°C for two hours, followed by incubation at 37°C for an additional two hours in order to anneal the primer to the template. 3'-5' exonuclease assay on 3' [3H] end-labeled DNase I- activated calf thymus DNA. The reaction mixtures (0.05ml) contained 50 mM Tris-HCl (pH 8.5), 4 mM MgC12, 20 mM DTT, 200 mM KCl (unless otherwise indicated), 400 ug/ml of bovine serum albumin, 55 NM [3H] end-labeled DNase I-activated calf thymus DNA 62 and DNA polymerase 7 Fr VI. Incubation was for 30 minutes at 30°C. The exonuclease reaction was terminated by the addition of 1 ml (j) t t. 10% trichloroacetic acid containing 0.4 M sodium pyrophosphate. The DNA was collected by filtration on Whatman GF/C glass fiber discs. The filters were washed with 10 ml of 1 M HCl containing 0.1 M sodium pyrophosphate. After drying, the filters were counted using a toluene-based scintillant. 3'-5' exonuclease assay on natural DNA and product analysis by denaturing gel electrophoresis . The reaction mixtures (0.05 ml) contained 50 mM Tris-HCl (pH 8.5), 4 mM MgClz, 20 mM DTT, 120 mM KCl, 400 ug/ml of bovine serum albumin, 4 uM DNA (the exonuclease substrate described above) unless otherwise indicated, and DNA polymerase 7 Fraction V, VI or VII. Incubation was at 30°C for 20 - minutes unless otherwise indicated. The exonuclease reaction was terminated by incubation for 10 min at 65°C in the presence of 1% sodium dodecyl sulfate and 10 mM EDTA. The samples were precipitated with ethanol in the presence of 10 ug of tRNA as carrier. The DNA pellet was then resuspended in 30 pl of dye mix (99% formamide, 0.1% bromophenol blue, 0.1% xylene cyanol), loaded onto an 18% polyacrylamide gel (0.75 mm) in the presence of 7 M urea and electrophoresed at 600-800 V for 4-6 hours. The gel was washed for 25 min in 15% glycerol and exposed to Kodak XAR film at -80°C using a Dupont Quanta III intensifying screen. Product analysis was performed by cutting and weighing the peaks produced from densitometric scanning of the autoradiograms using a Hoefer GS3OO scanning densitometer. Steady-state kinetic analysis of templateeprimer affinity of 63 the 3'-5' exonuclease. DNA polymerase y Fr.VI was assayed for 3'- 5' exonuclease activity using the M13 exonuclease substrate containing a 3'-terminal dAMPszMP mispair, as described above. Three or more determinations were made at each of eight DNA concentrations over a range of 0.1 to 16 uM. The data were analyzed by an enzyme kinetics program (EDATA v1.1, 1985 EMF Software) in an IBM XT computer; the kinetic values are presented as means. Nucleoside monophosphate inhibition of the 3'—5' exonuclease. DNA polymerase Y'Fr.VI was examined for 3'-5' exonuclease activity using the M13 exonuclease substrate containing a 3'-terminal dAMP:dAMP mispair as described above. Three or more determinations were made at each of four DNA concentrations over a concentration range of 0.25 to 2.5 uM and five AMP concentrations over a concentration range of 0 to 2.4 mM. The data were analyzed by an enzyme kinetics program (EDATA v1.1, 1985 EMF Software) in an IBM XT computer; the kinetic values listed are presented as means. The data presented as a Dixon analysis was analyzed by a linear regression program (Statworks v1.0 1986) in an Apple Macintosh computer. Primer extension analysis in the presence or absence of a 3'- terminal mispair. The reaction mixtures (0.05 ml) contained 50 mM Tris-HCl (pH 8.5), 4mM MgClz, 20 mM DTT, 120 mM KCl, 400 ug/ml of bovine serum albumin, the appropriate deoxynucleotides, 4 uM DNA (the M13 exonuclease substrate described above containing a 3'terminal dGMP:dGMP or dGMP:dAMP mispair or a 3'-terminal dAMP:dTMP basepair), and DNA polymerase y Fr.VI. Product analysis 64- was performed by denaturing gel electrophoresis and densitometric scanning of the autoradiograms as described above. Trapping experiment to detect transfer from the 3'-5' exonuclease site to the DNA polymerase site. DNA Polymerase 7 was preincubated for 20 minutes in a reaction volume of 0.7 ml at 30° C in the presence of 50 mM Tris-HCl (pH 8.5), 4 mM MgC12, 20 mM DTT, 120 mM KCl, 400 ug/ml of bovine serum albumin, and 4 uM single— stranded M13 DNA to which was annealed a 15 nucleotide primer to generate a 3'-terminal dGMP:dGMP mispair. The preincubation was followed by the simultaneous addition of dGTP, dTTP, and dATP (30 uM each) and DNase I-activated calf thymus DNA (180 uM). A control experiment was performed in the same manner omitting the 20 minute preincubation step. Here, a 17 nucleotide primer was annealed to the M13 DNA to generate a 3'-terminal dGMP:dGMP base pair. Aliquots (0.05 ml) were removed at the designated time point and precipitated with ethanol in the presence of 10 ug of tRNA as carrier. The resulting DNA pellet was resuspended in 30 ul of dye mix (99% formamide, 0.1% bromophenol blue, 0.1% xylene cyanol), loaded onto an 18% polyacrylamide gel (0.75 mm) in the presence of 7 M urea and electrophoresed at 600-800 V for 4-6 hours. The gel was washed for 25 minutes in 15% glycerol and exposed to Kodak XAR film at -80°C using a Dupont Quanta III intensifying screen. Product analysis was performed by cutting and weighing the peaks produced from densitometric scanning of the autoradiograms using a Hoefer GS3OO scanning densitometer. Preparation of embryo extract. All operations were performed at 0 - 4°C. D. melanogaster (Oregon R) embryos of average age 65 (9hr) were collected immediately before use and resuspended at a ratio of 4 ml/gm in 25 mM Hepes (pH 8.0), 10% glycerol, 0.3 M NaCl, 1 mM EDTA, 1 mM DTT, 10 mM sodium metabisulfite, 2 ug/ml leupeptin, and 2% sodium cholate and homogenized by 8 strokes in a 7 ml glass homogenizer. The homogenate was frozen immediately in liquid nitrogen and stored at -80°C. Preparation of a high-speed supernatant fraction. All operations were performed as described (191). This material was fractionated by the addition of 1.2 volumes of saturated ammonium sulfate, (pH 7.5) to achieve 55% of saturation at 0°C, and the suspension was incubated on ice for 2 hours. The precipitate was collected by centrifugation at 96,000 x g for 30 minutes at 3°C, and resuspended in 3.5 m1 of 10 mM phosphate buffer containing 20% glycerol. This fraction was dialyzed in 10 mM phosphate buffer containing 10% glycerol, 2 mM EDTA, 1mM PMSF, 10 mM sodium metabisulfite, 2 mM DTT, and 2 ug/ml leupeptin in a collodion bag (molecular weight cutoff 30,000) until an ionic equivalent of 150 mM NaCl was achieved. Protein determinations. Protein was detmined by the method of Bradford (192). Bovine serum albumin was used as the standard. Preparation of Antiserum. Antiserum directed against Drosophila DNA polymerase 7 was prepared with the near-homogeneous Fraction VI enzyme (159). A female New Zealand virgin white rabbit was immunized with 2.5 ug of DNA polymerase y in Freund's complete adjuvant by injection at or near the popliteal lymph node. Seven booster immunizations, each containing 2 ug of Fraction VI enzyme were administered. Bleedings were performed 8 to 10 days after 66 each boost. DNA polymerase and 3'-5' exonuclease inhibition studies. Prior to 3'-5' exonuclease assay, DNA polymerase Y(0.25 units) was preincubated on ice for 1 hour, in 50 mM Tris-HCl (pH 8.5), 4 mM MgC12, 20 mM DTT, 120 mM KCl, and 400 ug/ml of bovine serum albumin and the indicated amount of pre-immune or immune serum. Prior to DNA polymerase assay the enzyme was preincubated on ice for 1 hour in the same buffer including all four deoxynucleoside triphosphates at a concentration of 30 uM. Following preincubation, 40 ul aliquots were transferred to microfuge tubes containing 10 ul of 5'-end labeled singly-primed M13 DNA (4 uM final concentration) with a 3'-terminal dAMPszMP mispair to assay for exonuclease activity. To assay for DNA polymerase activity, a singly-primed M13 DNA (4 HM final concentration) with a 3'-terminal dAMP:dTMP base pair and 6 uCi of [3H1TTP were added. Protein gel electrophoresis, transfer and immunoblotting. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) was performed according to Laemmli (193). The proteins were transferred to nitrocellulose membranes (0.45 pm, Schleicher and Schuell) using a Hoefer Transfor electrophoresis unit model T322 for 6 hours at 70V in 190 mM glycine, 25 mM Tris and 20% (v/v) methanol. The membranes were washed in 20 mM Tris-Cl (pH 7.5), 150 mM NaCl and 0.05% (v/v) Tween 20 (TEST) for 10 minutes and blocked for 2 hours in 10 mM Tris-Cl (pH 7.5), 150 mM NaCl and 5% (w/v) nonfat milk. The membranes were probed with anti-DNA polymerase 7 serum (1:1000 in 20 mM Tris-Cl pH 7.5, 150 mM NaCl (TBS) and 0.5% nonfat milk) for 6 to 10 hours and 67 then washed three times for 10 minutes each in TBST and once for 5 minutes in TBS. After the washes were complete, a 2.5 hour 0 incubation was performed with protein A-alkaline phosphatase conjugate (1:2000, (v/v) or [125I] protein A (2 uCi/20 ml) in TBS (pH 8.0). The protein A-alkaline phosphatase treated membranes were washed three times for 10 minutes each in TBST, once for 5 minutes in TBS and once for 5 minutes in 100 mM NaCl, 5 mM MgC12 and 100 mM Tris, pH 9.5 (alkaline phosphatase buffer). The membranes were developed by incubation in alkaline phosphatase buffer containing nitroblue tetrazolium (330ug/ml) and 5-bromo-4- chloro-B—indolyl-phosphate (165ug/ml). The [1251] protein A-treated membranes were washed three times for 10 minutes each in TBST and exposed to Kodak XAR-5 film using a Dupont Quanta III intensifying screen. Glycerol gradient sedimentation under native conditions. DNA polymerase 7 (Fraction V) was sedimented into 12—30% glycerol gradients essentially as described by Wernette and Kaguni (22). Aliquots (0.02 ml) of the resulting fractions were removed and stabilized with an equal volume of buffer containing 20 mM KPO41fii 7.6, 80% glycerol, 0.015% Triton X-100, 2 mM EDTA, 2 mM DTT, 10 mM sodium metabisulfite, and 2 ug/ml leupeptin. The fractions were stored at -20°C and subsequently assayed for exonuclease and DNA polymerase activity. Glycerol gradient sedimentation under denaturing conditions. DNA polymerase 7 [Fraction VI (10.2 units,130 ng)] was diluted 1:20 in 0.67 ml 50% ethylene glycol, 20 mM KP04, 2 mM EDTA, 200 mM (NH4)2804, 0.015% TX-100, 1 mM PMSF, 10 mM sodium metabisulfite, 2 68 mM DTT, and 2 ug/ml leupeptin and incubated on ice for 2 hours, and then sedimented onto a 2-12% (v/v) glycerol gradient (10.6 ml) containing 50% ethylene glycol, 20 mM KP04, 2 mM EDTA, 200 mM (NH4)ZSO4, 0.015% TX-lOO, 1 mM PMSF, 10 mM sodium metabisulfite, 2 mM DTT, and 2 ug/ml leupeptin. Centrifugation was performed in a Beckman SW41 rotor at 3°C for 142 hours at 40,000 rpm. Sixty fractions (160 pl) were collected in silanized 1.5 ml microfuge tubes and analyzed for DNA polymerase and 3'-5' exonuclease activity and subunit composition. DNA polymerase I and bovine carbonic anhydrase were sedimented under the same conditions. Gradient fractions were analyzed for DNA polymerase activity (Pol I) and protein concentration by the Bradford assay (bovine carbonic anhydrase). Alternatively, DNA polymerase 1 [Fraction VI (57 units, 700 ng)] was incubated on ice for 4 hours in 0.25 ml containing 50 mM KPO4, 2 mM EDTA, 100 mM (NH4)2304, 0.015% TX-lOO, 1 mM PMSF, 10 mM sodium metabisulfite, 2 mM DTT, 2 ug/ml leupeptin and guanidine hydrochloride (Gu-HCl) at a final concentration of 1.2 M and then sedimented into a 12-30% glycerol gradient containing 0.8 M Gu— HCl, 20 mM KP04, 2 mM EDTA, 200 mM (NH4)2504, 0.015% TX-lOO, 1 mM PMSF, 10 mM sodium metabisulfite, 2 mM DTT, and 2 ug/ml leupeptin. Centrifugation was performed in a Beckman SW50.1 rotor at 3°C for 25 hours at 50,000 rpm. Fractions (0.08 ml) were collected in silianized 1.5 ml microfuge tubes and analyzed for DNA polymerase and 3'-5' exonuclease activities, and by immunoblot analysis. DNA polymerase I (25 units) and bovine carbonic anhydrase (700 pg) were sedimented under the same conditions. Gradient fractions were 69 analyzed for DNA polymerase activity (Pol I) and protein concentration by the Bradford assay (bovine carbonic anhydrase). Sephacryl S-200 gel filtration chromatography under denaturing conditions. DNA polymerase 7 [Fraction VI (33 or 66 units, 400 or 800 ng)] was incubated on ice for 1 hour in in 0.5 ml containing 20 mM KP04, 2 mM EDTA, 50 mM (NH4)ZSO4, 0.015% TX-100, 10 mM sodium metabisulfite, 2 mM DTT, 2 ug/ml leupeptin and Gu-HCl at final concentrations of 0.4, 1, or 1.5M. The samples were chromatographed at a flow rate of l/lO column volume per hour on a Sephacryl S-200 column (0.7 x 28 cm) equilibrated with 20 mM KP04, 2 mM EDTA, 50 mM (NH4)2804, 0.015% TX-100, 10 mM sodium metabisulfite, 2 mM DTT, 2 ug/ml leupeptin and Gu-HCl at final concentrations of 0.4, 1, or 1.5M. Fractions (0.1 ml) were collected and analyzed for DNA polymerase and 3'-5' exonuclease activities and by immunoblot analysis. DNA polymerase I (30 units) and bovine carbonic anhydrase (1mg) were chromatographed under the same conditions. Column fractions were analyzed for DNA polymerase activity (Pol I) and protein concentration by the Bradford assay (bovine carbonic anhydrase). Elution of proteins from sodium dodecyl sulfatetpolyacrylamide gels. One microgram each of DNA polymerase I and DNA polymerase 7 (Fraction VI) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis, protein elution from the polyacrylamide gel and enzyme renaturation as described by Hager and Burgess (194) with the following modifications. Visualization of the polypeptide of interest was achieved by staining for 10 minutes in 0.05% Coomassie blue in H20, followed by destaining with ‘70 H20 for 60 minutes. Protein elution from the gel was for 16 to 24 hours. Following acetone precipitation, the protein pellets were allowed to dry for 10 minutes and dissolved in 5 pl of 6 M Gu-HCl, 20% glycerol, 2 mM EDTA (pH 8.0), 200 mM (NH4)2804 (pH 7.5), 0.015% TX-100, 50 mM Tris-Cl (pH 7.9), 2 mM DTT, and 0.1 mg/ml B- lactoglobulin. The pellets were allowed to stand at room temperature for 20 minutes, and then diluted 50-fold with 20% glycerol, 50mM KPO4 (pH 7.6), 2 mM EDTA (pH 8.0), 200 mM (NH4)ZSO4 (pH 7.5), 0.015% T-XlOO, 50 mM Tris-Cl (pH 7.9), 2 mM DTT, and 0.1 mg/ml B-lactoglobulin. The proteins were allowed to renature for 16 to 24 hours on ice. DNA polymerase activities were determined by DNA polymerase assay on DNase I-activated calf thymus DNA. The Y’polymerase 3'-5' exonuclease activity was determined using single-stranded M13 DNA to which was annealed a S'end-labeled primer to yield a 3'-terminal mispair. Protein recovery for DNA polymerase I was determined by SDS-PAGE, Coomassie blue staining and densitometric scanning. Protein recovery for DNA polymerase 7 was determined by immunoblotting and comparison to known quantities of the y'polymerase following densitometric scanning of the immunoblot. Immunoprecipitation of Drosophila_7fipolymerase. DNA polymerase 7 (Fraction III, 19.7 units; Fraction I, 25 or 50 units, 300 or 600 mg equivalent of Drosophila embryos), embryo extract (48 or 96 mg equivalent of embryos) and the ammonium sulfate fractionated cytosolic S-100 fraction (450 or 900 mg equivalent of embryos) were diluted 1:1 with 10 mM NaPO4 (pH 7.0) and 154 mM NaCl (PBS) and incubated with preimmune serum or polyclonal antiserum overnight on 71 ice. Immune complexes were precipitated by incubation with 60 p; of preswollen protein A-agarose for 2 hours with gentle rotation. The precipitate was collected by centrifugation and washed once with PBS/0.05% Tween (v/v), twice with PBS, suspended in Laemmli sample buffer, heated for 10 minutes at 85°C and re-centrifuged. The supernatant fraction was then subjected to immunoblot analysis as described above. Chemical cleavage of 2;gsgpbila,7fipolymerase. DNA polymerase 7 (Fraction VI, 3 pg) was subjected to SDS-PAGE, Coomassie blue staining and destaining and excision of the a- and B-subunits. The fractionated 125 and 35 kDa polypeptides were subjected to proteolysis with Nechlorosuccinimide as described by Lischwe and Ochs (195), and the in situ proteolytic products from the isolated a- and B-subunits were separated on 5 to 15% gradient polyacrylamide gels and stained with silver as described by Wray et al. (196). CHAPTER III A Mismatch-Specific 3'-5' Exonuclease Associated with the Mitochondrial DNA Polymerase from DrosOphila Embryos INTRODUCTION The high fidelity of chromosomal DNA replication in both prokaryotes and eukaryotes (10"9 to 10‘11 errors per base pair replicated) results from a combination of DNA synthesis and postreplicational repair processes (197). Inasmuch as DNA repair is apparently absent in mitochondria (166), accurate replication of the mitochondrial genome may rely solely on the replication apparatus of the organelle. The issue of replication fidelity in mitochondria is of particular interest because animal mitochondrial DNA evolves at a rate 5- to 10-fold greater than single copy genomic DNA (165,198). It is not clear whether an increased mutation rate or an increase in the rate of fixation of mutations is the major factor. The former may result from infidelity during DNA synthesis or from lack of DNA repair. We have shown previously that the near-homogeneous mitochondrial DNA polymerase from Drosophila melanogaster embryos is highly accurate in nucleotide polymerization on single-stranded DNA (167). Further, the Drosophila y'polymerase does not exhibit a differential affinity for any dNTP nor does it misincorporate ATP (199). This is in contrast to a low replication fidelity reported for enzymes from HeLa cells and human placenta and fibroblasts (200-202). On the other hand, DNA polymerase 7 (Pol y) from chick embryos has been 72 73 capable of detecting a spectrum of base substitution and frameshift mutations (153,203). Of the four classes of eucaryotic DNA polymerases (a, B, y, and 5) only the 5 polymerase has been shown to contain a 3'-5' exonuclease component comparable to the prokaryotic enzymes (204). The 3'-5' exonuclease functions during DNA synthesis to excise misincorporated nucleotides at the 3'-terminus of a nascent DNA chain, and contributes as much as a factor of 100 to the overall fidelity of DNA synthesis (170,205,206). While high molecular weight forms of the a polymerase have been found in association with a 3'-5' exonuclease (207,208), and the near-homogeneous replicative enzyme from Drosophila embryos contains a cryptic 3'-5' exonuclease (138), DNA polymerases B and 7 are generally devoid of such an activity. Recently however, Kunkel has shown that the y polymerase from chick embryos contains an associated proofreading activity (162). We show here that the near-homogeneous mitochondrial enzyme from Drosophila embryos possesses a potent 3'-5' exonuclease which copurifies quantitatively with the DNA polymerase, and under in vitro reaction conditions excises mismatched nucleotides at the 3'- end of a DNA primer prior to nucleotide polymerization. 74 RESULTS Drosophila DNA Polymerase ‘y does not catalyze detectable dNTP turnover during DNA polymerization. The mitochondrial DNA polymerase from embryos of Drosophila melanogaster consists of a 125 kDa DNA polymerase catalytic subunit and a 35 kDa subunit of unknown function (159). Because the enzyme replicates DNA with high fidelity (167), we examined the possibility that the Y polymerase contains a 3'-5' exonucleolytic editing function by assay of dCMP turnover. Drosophila DNA polymerase 7 exhibited no detectable turnover of dCTP even at high polymerase levels, under conditions in which E. coli DNA polymerase I and DNA polymerase III yielded substantial amounts of dCMP (Table 1). While the dNTP turnover assay allows direct measurement of exonucleolytic excision during polymerization, it measures removal of both correctly paired and mismatched bases. The data in Table 1 suggest that either the mitochondrial enzyme does not possess a 3'—5' exonuclease activity or, in contrast to E. coli DNA polymerases I and III, that this function might be highly mismatch specific under DNA polymerization conditions. In view of the high fidelity of nucleotide polymerization by P01 7, the removal of only mismatched nucleotides during DNA polymerization may not be detectable in the turnover assay. A 3'-5' Exonuclease activity associated with Drosophila mitochondrial DNA. polymerase excises 3'-terminal. mismatched nucleotides from synthetic DNA. To determine whether or not the Drosophila mitochondrial DNA polymerase contains a mismatch- specific 3'-5' exonuclease function, Pol 7 was assayed on the 75 Table 1. Measurements of dNTP turnover catalyzed by D. melanogaster DNA polymerase ‘y and by E. coli DNA polymerase I and DNA polymerase III holoenzyme E n z yme dCMP . pmol Incorporated Formed DNA polymerase Y 16.9 - <0.06 66.7 <0.06 185.8 <0.06 DNA Polymerase I 17.6 1.74 52.1 3.74 DNA Polymerase III 5.8 2.05 11.0 3.20 Reaction conditions were as described under = (dCMP formed/(dCMP incorporated + dCMP formed)) dCTP, % turnover* (0.35 <0.09 <0.03 26.1 22.5 "Methods." * % turnover x 100. 76 synthetic DNA substrate poly(dA)7oo-p(dT)15 containing either a 3'- terminal mismatched nucleotide ([3H ]dCMP or [3H]dAMP) or a correctly paired nucleotide ([3H]dTMP). Under non-polymerization conditions, DNA polymerase 7 exhibited hydrolysis of the 3'— nucleotide pairs dAMP:dCMP, dAMP:dAMP, and dAMP:dTMP at rates of 15.0, 35.8 and 1.05%, respectively, relative to the rate of nucleotide polymerization on poly(dA)7oo. p(dT)15 (Table 2). While E. coli DNA polymerase I showed a 2.5-fold greater rate of hydrolysis of the mispaired termini, mismatch specificity for both the dAMP:dCMP and dAMP:dAMP mispairs relative to the dAMP:dTMP pair was 2.5-fold lower than that observed with the Y polymerase. At the same time, no hydrolysis of poly(dA)-p(dT) in which the (dT)15 was 5'-end labeled with PrJZP] was detected with the mitochondrial enzyme (data not shown). Thus, the data demonstrate a 3'-5' exonuclease activity associated with Drosophila mtDNA polymerase which has a 14- to 34-fold specificity for A:C and A:A mispairs relative to the correctly paired 3'-nucleotide. Quantitative copurification of 3'-5' exonuclease and DNA polymerase. To demonstrate the quantitative association of the 3'-5' exonuclease with DNA polymerase y, co-chromatography of the two enzymatic activities was examined at each of three purification steps. DNA polymerase activity obtained upon octyl-Sepharose chromatography (Fraction IV) was further chromatographed on Cibacron blue agarose (Fraction V) and sedimented in a glycerol gradient as described in our standard purification scheme (159). Finally, the resulting near-homogeneous enzyme (Fraction VI) was subjected to gel 77 Table 2. 3'-5' Exonuclease activity of Drosophila DNA polymerase ‘y 3'-terminal nucleotide WM DNA polymerase Mismatchsd Matshfid activity, unit [3H]dC [3H]dA [3H]dT Enzyme DNA polymerase 7 0.12 21.3 39.3 0.9 Preparation 1 0.12 17.2 ND 1.2 0.15 23.2 57.6 2.1 Preparation 2 0.15 18.8 ND ND DNA polymerase I , 0.12 46.9 102 8.0 DNA polymerase 7 (Fraction VI) and E. coli DNA polymerase I were assayed for DNA polymerase activity and hydrolysis of terminal mismatched nucleotides as described under "Methods." Poly(dA)7oo- p(dT)1o was the substrate for nucleotide polymerization, and poly(dA)7oo-p(dT)15-[3H]dC1, -[3H]dA1, or -[3H]dT1 were the substrates for exonucleolytic hydrolysis. ND, not determined. 78 filtration on FPLC (fast protein liquid chromatography) Superose :2 (Fraction VII). These enzyme fractions were assayed for DNA polymerase and 3'-5' exonuclease activities in a time course analysis (Figure 9). In this case, the substrate used for exonuclease assay was a natural DNA to which was annealed a 5'-end labeled 3'-terminal mismatched primer of 15 nucleotides. Product analysis by denaturing gel electrophoresis and quantitation of the exonuclease assay were as described under "Methods." Both DNA polymerase and exonuclease activity were linear for 60 minutes at 30°C (Figure 9A). The ratios of DNA polymerase to 3'-5' exonuclease over the 60 minute time course were 14.5, 10.8 and 8.7 for enzyme Fractions V-VII, respectively. Thus, the ratio of the two enzyme activities varies only 20 percent between the near- homogeneous Fraction VI and Fraction VII preparations of DNA polymerase 1. The slightly higher DNA polymerase to exonuclease ratio in the enzyme from Cibacron blue agarose (Fraction V, 20% pure) suggests the presence of inhibitors of the 3'-5' exonuclease which were removed by subsequent glycerol gradient sedimentation. These were most likely contaminating endonuclease, 5'-3' exonuclease and/or phosphatase, as demonstrated by alternate assays (not shown). The data show the quantitative association of the 3'-5' exonuclease with the Drosophila mitochondrial DNA polymerase. Further, examination of the products of the exonuclease assay (Figure 9D) clearly demonstrate that the 3'-5' exonuclease is highly mismatch- specific, removing only the 3'-terminal mismatched nucleotide to yield products of 14 nucleotides in length. No detectable hydrolysis of a correctly paired 12-nucleotide primer present as a 79 Figure 9. Copurification of 3'-5‘ exonuclease and DNA polymerase. (A-C), A time course of DNA synthesis and exonucleolytic hydrolysis was performed as described under "Methods." DNA polymerase activity was assayed on DNase I-activated calf thymus DNA and 3'-5' exonuclease activity was assayed on M13 DNA to which was annealed a 5'-[32P]-labeled lS-nucleotide primer containing a 3'-terminal mismatched nucleotide (dAMP opposite a template dAMP). Nucleotide polymerized (open circles) and hydrolyzed (closed circles) are expressed in pmol. A: DNA polymerase 7 Fraction V; 3: Fraction VI; C: Fraction VII. D. Terminal mismatch excision was analyzed by gel electrophoresis. Products obtained at the time intervals of 0, 10, 20, 40 and 60 minutes indicated in A-C were isolated, denatured and electrophoresed in a denaturing 18% polyacrylamide gel as described under ”Methods.” Lanes 1-5: Fraction V; lanes 6-10: Fraction VI; lanes 11—15: Fraction VII. NUCLEOTIDE POLYMERlZED 92 3 O 80 C _ P -12 u— )— 1 F P d 8 I- .- d 4 L I- -( l 1 1 1 1 1 1 1 1 1 1 1 0 30 60 O 30 60 O 30 60 TIME , minutes I 2345 6 7 89l0 lll2l3l4l5 ‘- - ~ _ " . ’-|5 ” -‘o --4 ' . y...“ ‘ O -‘ . ‘ e... -- «4* Figure 9 NUCLEOTIDE HYDROLYZED, x IO' 81 contaminant in the substrate is observed, nor are other product lengths seen. Cosedimentation of 3'-5' exonuclease and DNA polymerase. To substantiate further the quantitative association of the 3'-5' exonuclease with Drosophila DNA polymerase y, a sedimentation analysis of the Fraction V enzyme under native conditions was undertaken. As shown in Figure 10, the DNA polymerase and exonuclease activities were precisely coincident. Further, the ratio of DNA polymerase to 3'-5' exonuclease across the peak fractions was invariant: the DNA polymerase to exonuclease ratios for the indicated fractions between 27 and 35 were 0.91, 0.90, 1.0, 0.94, 0.82, respectively, where the peak fraction (Fraction 31) was assigned an arbitrary value of 1.0. Again, the specificity of hydrolysis of the 3'-terminal mismatched nucleotide from the 15-nucleotide primer is demonstrated (Figure 108). An insignificant fraction of the correctly paired 12-nucleotide primer (dGMP:dGMP) is hydrolyzed to an ll-mer under conditions where hydrolysis of the mismatched 15-nucleotide primer (dAMP:dAMP) is 54 percent (Fraction 31). Drosophila DNA polymerase 7 exhibits exonucleolytic editing under conditions of nucleotide polymerization. The Drosophila mitochondrial DNA polymerase specifically hydrolyzes 3'- terminal mismatched nucleotides from synthetic and natural DNA substrates in the absence of DNA synthesis. To determine if exonucleolytic proofreading occurs under conditions of DNA synthesis, a primer extension analysis was performed. 82 Figure 10. Cosedimentation of 3'-5' exonuclease and DNA polymerase. A. DNA polymerase y Fraction V was sedimented under native conditions in a 12-30% glycerol gradient and assayed for DNA polymerase (open circles) and exonuclease (closed circles) activity as described under "Methods" and in the legend to Figure 9. Fractions 1 and 62 represent the bottom and the top of the gradient, respectively. Exonuclease activity is expressed as % of substrate hydrolyzed; the 54% hydrolysis indicated in fraction 31 corresponds to 6.5 pmol hydrolyzed per pl of Fr.VI enzyme. 3. Terminal mismatch excision was analyzed by gel electrophoresis as described under "Methods” and in the legend to Figure 9. POL 7 ACTIVITY, pmol/pl 83 8 I2 I8 24 SOIfli‘Wl48 54 60 FRACTION NUMBER Is-W l4- W:r“" I2 _ {.1 ¢-*“.*g¢ 3.. ’1'- J.,: r.-."— # "' .5 .52; 2322.25.53.35 seems-z FRACTION NUMBER Figure 10 A I20- EXONUCLEASE ACTIVITY, °/o 84 DNA polymerase 7 was incubated with an M13 DNA substrate containing a dGMP:dGMP mispaired primer under DNA polymerization conditions in the absence of dCTP, the next nucleotide required for primer extension after exonucleolytic hydrolysis. The time course analysis presented in Figure 11 (lanes 1-8) demonstrates that no detectable extension of the 3'-terminal mismatch occurs. Instead, substantial hydrolysis of the 3'-mispair is observed. Thus, exonucleolytic editing of the mispair is required to permit primer extension by Pol 7. If dATP is omitted instead of dCTP, proofreading DNA synthesis occurs without significant accumulation of the 14-mer product of the hydrolysis reaction (Figure 11, lanes 9-16). In fact, the rate of disappearance of the mispaired 15-mer is nearly identical to the rate of appearance of the product 19-mer (the product length expected for primer extension in the absence of dATP, Figure 12A). The data indicate that exonucleolytic hydrolysis is the rate limiting step in proofreading DNA synthesis. Further, the rate of exonucleolytic hydrolysis is lower during primer extension than that observed in the absence of DNA synthesis (Figure 12A). To compare proofreading DNA synthesis to simple primer extension by the mitochondrial DNA polymerase, a pre-incubation in the absence of dNTPs was followed by incubation under DNA polymerization conditions in the absence of dATP. The rate of extension of the correctly paired 14-mer, produced during the pre-incubation in the absence of DNA synthesis, was dramatically higher than that of the mispaired lS—mer (Figure 11, lanes 17-24): no significant 85 Figure 11. Proofreading DNA synthesis on 2413 DNA by DNA polymerasety. DNA polymerase y Fraction VI was incubated with an M13 exonuclease substrate (4 pM) containing a dGMP:dGMP terminal mismatch under standard DNA polymerase assay conditions with the exceptions indicated below. Terminal mismatch excision was analyzed as described under "Methods" at the time intervals of 0, 2, 4, 10, 20, 30 and 40 minutes. Lanes 1-8: dCTP was omitted from the incubation mixture. Lane 1 represents a no enzyme control, and lanes 2-8 correspond to the time intervals indicated above. The product length expected upon primer extension is 23 nucleotides. Lanes 9-16: dATP was omitted from the incubation mixture. Lane 9 represents a no enzyme control and lanes 10-16 correspond to the time intervals indicated above. The product length expected upon primer.extension is 19 nucleotides. Lanes 17-24: a preincubation for 12 minutes in the absence of dNTPs was performed prior to DNA synthesis in the absence of dATP. Lane 17 represents a no enzyme control and lanes 18-24 correspond to the time intervals indicated above. (-)dCTP 2345678 86 9|OII FIdATP :2 :3 I4 l5 I6 Figure 11 I-) dATP. (+I pro-inc I7 l8 :9 20 2| 22 23 24 -22 87 utilization of the mismatched primer occurred until most of the paired primer was extended to 19-mer product. Preferential utilization of the paired versus mispaired primers suggests a mechanism in which P01 7 dissociates more readily from mispaired primers prior to hydrolysis than from paired primers prior to polymerization. In any case, quantitation of the rate of disappearance of the correctly paired (dGMP:dGMP) and mispaired primers (dGMP:dGMP) during the linear phase of each reaction revealed a 4.5-fold greater rate of extension of the former relative to the latter,(Figure 12B) providing further evidence that mismatch hydrolysis is the rate limiting step in proofreading DNA synthesis. 88 Figure 12. Time course of utilization of paired versus mispaired primers on M13 DNA as described under "Methods." A: Rate of hydrolysis of the mispaired lS-mer in the absence of dCTP (no primer extension), (open circles); rate of hydrolysis of the mispaired 15-mer in the absence of dATP (open triangles) (hydrolysis during proofreading), (closed circles); rate of primer extension in the absence of dATP (product formation during proofreading). 8: Rate of primer extension of the correctly paired 14—mer in the absence of dATP following preincubation in the absence of all dNTPs, (open circles); rate of proofreading DNA synthesis in the absence of dATP following preincubation, (closed circles). 89 .A _ _ . w w m m s. .8535 SEE 4o 30 20 IO minutes TIME , Figure 12 90 DISCUSSION Replication of chromosomal DNA is, by necessity, a highly faithful process: the maintenance of genetic integrity is dependent on accurate DNA replication, and infidelity may relate to aging and disease (197). On the other hand, mitochondrial DNA is present in several thousand copies per cell (thereby rendering individual molecules dispensible), and exhibits a significantly higher rate of evolution (165,198). Nevertheless, mitochondrial genomes from organisms as divergent as Drosophila and man have an identical gene content and highly similar gene organization (12,14,209). We have shown previously that DNA polymerase 7 from Drosophila embryos is highly accurate in the synthesis of DNA relative to replicative enzymes from both prokaryotic and eukaryotic sources (167). Here we have further examined its catalytic properties to begin to elucidate the mechanism by which this mitochondrial enzyme achieves high fidelity in nucleotide polymerization. Our data ‘indicate that the two subunit Drosophila y polymerase contains a mismatch-specific 3'-5' exonuclease activity which most likely provides a proofreading function during mtDNA replication. In contrast to other DNA polymerases containing an editing function, Drosophila Pol 7 shows no detectable dNTP turnover under optimal conditions for DNA synthesis. Under such assay conditions, E. coli DNA polymerases I and III exhibit dNTP turnover values of ~7.0% (210, this study) and 23% (211, this study), respectively. Likewise, turnover values of 80-90% and 34% were reported for eukaryotic 5 polymerase (204) and the a-subunit of Drosophila a polymerase (210), respectively. Because all of these enzymes 91 misincorporate nucleotides at frequencies of <10‘5 per base pair replicated, most of the dNMP formed either results from hydrolysis of correctly paired nucleotides or from template dependent hydrolysis of nucleotides which have not actually been incorporated. The latter possibility may result either from conformational or from kinetic proofreading during the base selection step prior to nucleotide polymerization (197). Assuming that the optimal salt concentration for polymerization (200 mM KCl) does not inactivate the exonuclease, it appears that Drosophila y'polymerase may specifically hydrolyze only misincorporated nucleotides and/or have a more accurate mechanism for nucleotide selection. Although dNTP turnover was not detected, the Drosophila mitochondrial DNA polymerase catalyzed substantial hydrolysis of 3'- terminal mispaired nucleotides from synthetic DNA substrates, exhibiting an exonuclease to polymerase activity ratio 2.5-fold lower than that of E. coli DNA polymerase I. Exonucleolytic excision was in the 3'-5' direction only and was highly specific for the mismatched pairs dAMP:dCMP and dAMP:dAMP relative to the correct dAMP:dTMP base pair (14- and 34-fold, respectively). As with the Drosophila DNA polymerase a (138), the exonuclease component of DNA polymerase Y shows greater activity in hydrolysis of an A:A mispair relative to an dAMP:dCMP mispair. If uncorrected, an dAMP:dAMP mispair would yield an A-T transversion, the most frequent third position codon change in four-codon families of Drosophila mitochondrial protein genes (212). Because the exonuclease function of the mitochondrial DNA polymerase is particularly effective in hydrolyzing this mispair, it may be more likely that the anomalously 92 high frequency of A-T substitutions in Drosophila mtDNA results from a strong selection in favor of fixation of the products of such mutations. It is notable, however, that the specificity and efficiency of exonucleolytic editing and its effect on DNA replication fidelity in vitro is dependent on both the template and the mispair being examined, and on reaction conditions (197,213). The specificity of Drosophila mitochondrial DNA polymerase in hydrolysis of a noncomplementary 3'-terminus is also demonstrated on natural DNA substrates. The product analysis experiments indicate that Pol y excises only the mispaired nucleotide (dAMP:dAMP or dGMP:dGMP) to yield products diminished in length by a single nucleotide: there was no significant hydrolysis of either an dAMP:dTMP or a dGMP:dGMP pair at the 14-nucleotide position, or of a dGMP:dGMP pair at the 3'-terminus of a contaminating primer of 12- nucleotides. In a similar analysis, the exonuclease associated with the mitochondrial DNA polymerase from chick embryos was not highly specific but showed a 2- to 6-fold preference for mispaired termini (162). Under DNA polymerization conditions, Drosophila DNA polymerase 7 does not extend a mismatched 3'-terminus: in the absence of the next nucleotide required after mismatch-hydrolysis, only the hydrolysis step is observed. The data indicate that exonucleolytic hydrolysis is the rate limiting step in proofreading DNA synthesis. Further, the rate of the hydrolytic step is slower under polymerization relative to non-polymerization conditions, suggesting that Y polymerase does not dissociate after the hydrolytic step but utilizes the correctly paired product terminus for subsequent 93 nucleotide polymerization. Finally, Pol 7 extends a complementary 3'-OH terminus, under these reaction conditions, at at least a 4- fold greater rate than a mispaired 3'-terminus. As observed in Figures 11 and 12, the rate of exonucleolytic hydrolysis under non- polymerization conditions is approximately one-third of the rate of DNA polymerization; the slower rate of extension of the mispaired primer most likely reflects the time required for the enzyme to excise the noncomplementary nucleotide. Taken together, the data indicate that Drosophila Pol y'proofreads errors during in vitro DNA synthesis. The effect of the exonuclease function on DNA replication fidelity remains to be determined directly. Finally, we have demonstrated that the mismatch-specific 3'-5' exonuclease quantitatively copurifies with the two subunit Drosophila y polymerase. The peaks of the two activities were coincident upon glycerol gradient sedimentation of the Fraction V enzyme. Further, the DNA polymerase to exonuclease activity ratios remained nearly constant upon gel filtration of the near-homogeneous Fraction VI enzyme, as does the 1:1 stoichiometry of the two subunits (159). Experiments to determine the subunit association of the 3'-5' exonuclease activity are underway. CHAPTER IV Coordination of 3'-5' Exonuclease and DNA Polymerase Function in Drosophila y Polymerase INTRODUCTION Accurate in vivo DNA synthesis is ensured by a multistep process. The first step involves the selection and subsequent incorporation of nucleotides complementary to the template DNA strand by DNA polymerase. This, in conjunction with post- replicational repair allow for an average in vivo error frequency of 10’9 to 10‘11 errors per base pair replicated (214). The AG of discrimination between correct and incorrect Watson and Crick base pairs is only sufficient to provide a fidelity of one error per 102 to 103 nucleotides incorporated (215). However, the in vitro error rate of E. coli DNA polymerase I is one in 105 to 106 base pairs replicated (216,217). Further, the in vitro error rates of the eukaryotic a, y, and 6 polymerases are 1/118,000, 1/260,000 and l/470,000, respectively (142,162,151). Thus, faithful DNA replication, though essential, remains only partially explained. An important component of the DNA replication machinery is proofreading at each nucleotide addition by a 3'-5' exonuclease (169). Proofreading exonucleases have been shown to possess the following characteristics: i) they function coordinately with the DNA polymerase; ii) they are physically associated with the DNA polymerase; iii) they preferentially excise mispaired as Opposed to paired 3'— primer termini; iv) they digest single- 94 95 stranded DNA; and v) they are generally rendered less effec2: e by a nucleotide pool bias and by the presence of nucleoside monophosphates (169,218). Proofreading exonucleases have been shown to be associated with the bacteriophage T7, T4 and T5 DNA polymerases as well as E. coli DNA polymerase I and DNA polymerase III holoenzyme (172,219,220,221,222). With regard to the five previously defined classes of eukaryotic DNA polymerases, only the 5 polymerase was shown to have an associated 3'-5' exonuclease comparable to that of the prokaryotic DNA polymerases (204). DNA polymerase a, as isolated from Drosophila, has been shown to possess a cryptic, mispair specific and potent 3'-5' proofreading exonuclease associated with the 182 kDa polymerase catalytic subunit (138). Further, 3'-5' exonucleases are apparently associated with higher molecular weight forms of the a polymerase (174,207,208). More recently, the quantitative association of a 3'-5' exonuclease activity has been demonstrated for the mitochondrial DNA polymerases isolated from chick embryos, porcine liver, Drosophila embryos and Xenopus laevis oocytes (161,162,163,164). In addition, the 3'-5' exonucleases associated with the y polymerases from Drosophila embryos, chick embryos and porcine liver have been shown to function as proofreading exonucleases in vitro. A relatively undefined segment of mitochondrial DNA (mtDNA) replication is fidelity. This is of particular interest because the evolutionary rate of animal mtDNA is 5- to 10-fold greater than that of single copy genomic DNA (165,196). In addition, 96 the mitochondrial genome is saturated with protein encoding sequences: the small 14.5 to 19.7 kb genome (in animal cells. encodes 13 proteins required for oxidative phosphorylation a: $1 electron transport. Further, this genome encodes its own protein synthetic machinery (69). These features dictate a requirement for accurate DNA replication. Previously, the high rate of evolution of animal mtDNA was thought to be the result of inaccurate mtDNA replication -- an unfaithful y polymerase -— and/or the lack of a mtDNA repair system (166). Recent Studies indicate that the Y polymerases isolated from chick embryos and porcine liver replicate DNA with high fidelity (161,162). Further, the Drosophila y polymerase has been shown to replicate DNA with an accuracy equal to E. coli DNA polymerase III holoenzyme and Drosophila DNA polymerase a (167). Thus, the anomaly between a high mtDNA evolutionary rate and a highly accurate mtDNA polymerase, may indicate that the major contributor to the high evolutionary rate is an increased rate in the fixation of errors. This is most likely a direct effect of the apparent absence of a mtDNA repair mechanism. The ability of a DNA polymerase and proofreading exonuclease to function coordinately lends evidence that the in vivo function of the exonuclease is to enhance the overall fidelity of DNA synthesis. In this study we investigated the functional coordination of the 3'-5' exonuclease and DNA polymerase activities associated with the Drosophila mtDNA polymerase. We examined: i) the enzymatic and kinetic properties of the 3'—5' exonuclease, ii) differential mispair excision, iii) mispair 97 specificity of exonucleolytic excision and iv) the effe ts of nucleoside monophosphates and an over abundance of the next correct nucleotide on mispair excision and primer extension. Further, evidence is presented suggesting that the mechanism of template-primer transfer from the 3'-5' exonuclease active site to the DNA polymerase active site is intermolecular. 98 RESULTS Monovalent cation optima for DNA polymerase and 3'-5' exonuclease activities on singly-primed M13 and DNase I- activated calf thymus DNA are similar. The DNA polymerase and 3'-5' exonuclease activities of V 0 polymerase require KCl to exhibit maximal reaction rates. To demonstrate a functional coordination of the two enzyme activities, their KCl optima were determined using two DNA substrates varying in primer density. The optimal KCl concentration for both the DNA polymerase and 3'-5' exonuclease on singly-primed M13 DNA (containing a 3'-terminal dAMP:dAMP mispair in the exonuclease assay) is ~140 mM (Fig 13). Furthermore, the DNA polymerase and 3'-5' exonuclease activity profiles are coincident. No deviation in the 3'—5' exonuclease activity with regard to mispair specificity was observed at suboptimal KCl concentrations (data not shown). Previous work in this laboratory revealed a lack of dNTP turnover during DNA polymerization by the Drosophila mtDNA polymerase (163). It had been suggested that the absence of dNTP turnover might be attributed to the presence of 200 mM KCl rendering the 3'-5' exonuclease inactive. To address this issue and to determine the optimal KCl concentration of the 3'-5' exonuclease using a DNA substrate with high primer density possessing 3'-terminal mispairs (see "Methods" and legend to Figure 14), a titration experiment was performed (Fig 14). Maximal DNA polymerase and 3'-5' exonuclease activity are achieved at 200 mM and 240 mM KCl, respectively. The activity 99 Figure 13. Dependence of DNA polymerase and 3'-S' exonuclease activities on monovalent cation concentration in the presence of a singly-primed DNA substrate. DNA polymerase 7 Fr. VI was assayed for DNA polymerase activity (0.08 units, closed circles) and 3'-S' exonuclease activity (0.17 units, open circles) on singly-primed M13 DNA as described under ”Methods,” in the presence of the indicated KCl concentrations. 100 20 20 w .oeuaaowpam seemed: 0 1 . F 1 ooueuoduooeu maze seam - 5 0 5 o 1 250 200 150 100 50 KCl, Figure 13 101 Figure 14. Dependence of DNA polymerase and 3'-S' exonuclease activities on monovalent cation concentration in the presence of a DNA substrate with high primer density. DNA polymerase 7 Fr. VI was assayed for DNA polymerase activity (0.11 units, closed circles) and 3'-5' exonuclease activity (0.25 units, open circles) on DNase 1- activated calf thymus DNA as described in "Methods," in the presence of the indicated KCl concentrations. 102 a .powaaoupam “domed: 4 3 2 1| 0 — P n P D - a - .7. ~ ~ 1 C v l O ‘ b\“\ A \ \ \ I. \\ kw \\ 4 b\ \ l \\ C O ’ l O""' I, I l O I . Oi} "b D, . nu I. I — p — p — p - p 0 0 0 0 0 5 4 3 2 1 o oouewomwoosu maze «can 100 150 200 250 300 350 400 450 500 50 KCl, Figure 14 103 profiles are nearly coincident and the difference between the 3 '-5' exonuclease activity between 200 mM and 240 mM KCl is less than 10%. This is also true for the DNA polymerase. Taken together, the data indicate nearly identical monovalent cation optima (KCl) for the DNA polymerase and 3'-5' exonuclease on the two DNA substrates examined, which vary significantly in prime: density. Divalent cation optima for the DNA polymerase and the 3'-5' exonuclease are similar. DNA polymerase 7, like all DNA polymerases, requires Mg2+ for activity. Titration of MgClz using singly-primed M13 DNA as the template-primer reveals similar divalent metal cation optima for DNA polymerase and 3'-5' exonuclease activities (Fig. 15). Both 3'-5' exonuclease and DNA polymerase activities increase Sigmoidally from 0.1 to 0.5 mM and 0.25 to 1.0 mM MgC12, respectively. MgClz exhibits its optimal effects at approximately 0.5 and 1.0 mM for 3'-5' exonuclease and DNA P°lymerase, respectively. The 3'-5' exonuclease exhibits 96% of its maximum activity at 1.0 mM MgC12. Likewise, the DNA p°lymerase was >90% active at 0.5 mM MgC12 relative to the Optimal activity achieved at 1 mM MgC12. Thus, the data suggest nearly identical Mg2+ optima for the two enzyme activities associated with the Y polymerase. Notably, MgClz concentrations °f 8 mM versus 15 mM were required to reduce the 3'-5' e"‘Omlclease and DNA polymerase activities by half, indicating a di . . . . fferentlal tolerence to increaSing MgClz concentrations. 104 Figure 15. Dependence of DNA polymerase and 3'-5' exonuclease activity on divalent metal cation concentration in the presence of a singly-primed DNA substrate. DNA polymerase 7 Fr. VI was assayed for DNA polymerase activity (0.18 units, closed circles) and 3'-5' exonuclease activity (0.09 units, open circles) on singly-primed M13 DNA as described under ."Methods, " in the presence of the indicated MgClz concentrations. % Relative Activity, 105 100 f". '1 5' 100 90 'i 80 I, 5: 80 I, ' 33 6° 1 , O 40 70; 3. 20 u ‘3 . U H 0 a I a l a l L 6° .2 0.0 1.0 2.0 3.0 4.0 I NgCl Concentration, ml! 50 I e 2 40 I 30 0 I 20 [I O 10 - o 4 a l I l O 5 10 15 20 25 30 MgCl2 Concentration, m Figure 15 106 Template-primer affinity of 3'-5' exonuclease The Km and kcat for DNA polymerase activity on a singly- primed natural DNA substrate have been determined previously (167) . To evaluate further the functional coordination of the two enzyme activities associated with Drosophila y polymerase, the Km and kcat for the 3'-5' exonuclease were determined on the singly-primed M13 exonuclease substrate (Table 3) . The Km values determined for 3'-5’ exonuclease and DNA polymerase are 1.38 pM and 1.06 pM, respectively. Thus, the template—primer affinities for the two enzyme activities are nearly identical. However, the kcat of the DNA polymerase on singly-primed DNA is approximately 1700-fold greater than the kcat for the 3'-5' exonuclease . Inhibition of 3'-5' exonuclease by 5'AMP Several investigators have shown that nucleoside m°n°ph<33phates (NMPs) inhibit proofreading 3'-—5' exonucleases I7°r162,l65,224,225) . A study performed by Que, Downey and So TMP functions as a competitive inhibitor of the single-stranded pOlY dT used for nucleotide hydrolysis by the 3'-5' exonuclease associated with E. coli DNA polymerase I Klenow fragment (218). Further, NMP inhibition has been used as a technique to determine the relative contribution of the 3'-5' exonuclease associated with a DNA polymerase to the fidelity of DNA Synthesis (151,162,169,234) . We examined the ability of 5'AMP to inhibit the 3'-5' exonuclease associated with the DrosOphila 7 pOlymerase, in an analysis designed to 107 Table 3. Template-primer affinity of the 3'-5' exonuclease and DNA polymerase associated with the Drosophila ‘y polymerase for singly-primed DNA. Enzyme Activity Km kcat (pM) nt incorp/hydrolyzed s‘l enzyme molecule‘1 3'-5' exonuclease 1.38 +/- 0.39 2.37 x 10‘4 *DNA polymerase *1.06 +/- 0.27 0.396 * Taken from: Wernette, C.M. Conway, M.C. and Kaguni, L.S. (1988) Biochemistry 27, 6046-6054 The exonuclease assay was performed as described in "Methods" on an M13 DNA containing a 3'-terminal dAMP:dAMP mispaired primer. 108 discern the mode of inhibition by AMP relative to the M13 dAMP:dAMP 3’-terminal mispaired template-primer. AMP was shown to be the most appropriate inhibitor in a survey of NMPs (data not shown). Concentrations of the template-primer were chosen to surround the Km to optimize detection of the effects of AMP on 3'-5' exonuclease activity. The data in Table 4 indicate that the 3'-S' exonuclease is inhibited in a concentration dependent manner with regard to both AMP and template-primer termini. This is evidenced by increased inhibition of the 3'-5' exonuclease with increasing AMP concentration (constant DNA concentration) and with decreasing template-primer concentration (constant AMP concentration). The amount of inhibition by AMP does not exceed 50% even in the presence of 2.4 mM AMP and a DNA concentration of 0.25 pM, the most stringent conditions examined. Under the above conditions, the ratio of AMP to DNA is 10000:1. Perhaps more significantly, the ratio of AMP to available 3' termini is ~ 108:1. Dixon analysis and replotting of Km vs I and l/Vmax vs I following Lineweaver-Burk analysis was performed to determine the Ki of AMP and the mechanism of inhibition (Table 5). Dixon analYSis provided a Ki for ~3 mM and evidence indicating that AMP funotions as a competitive inhibitor for template-primer termini. Replotting of Km vs I and 1/Vmax vs I rendered respective Ki's for AMP of 4 mM and 26 mM, indicating AMP is a C°mPEtitive inhibitor. Thus, AMP is a relatively poor inhibitor of the 3'-5' exonuclease associated with 901 Y- \II e-‘v 109 Table 4. Inhibition of the 3'-5' exonuclease activity associated with the Drosophila y polymerase by adensosine S'monophosphate (AMP). AMP ' ' ' " ' ° (NM) DNA (pM): 0.25 0.5 1.0 2.5 o __ __ __ __ 160 10.7 7.5 5.9 5.7 560 32.1 20.7 11.6 10.6 1600 . 35.5 30.3 24.4 17.8 2400 47.9 45.1 40.4 37.3 The exonuclease assay was performed as in Table 3. 110 Table 5. Ni for AMP in mispair excision by the 3'- 5'exonuclease associated with Drosophila y polymerase: Dixon. and. Lineweaver-Burk analyses Analysis DNA concentration Ki of AMP (QM) 1mM) Dixon 0.2 3.0 0.5 2.8 1.0 3.0 2.5 2.9 Lineweaver-Burk . Replot Km vs I 4 l/Vmax vs I 2 6 111 A similar analysis was performed with E. coli DNA polymerase I Klenow fragment (Pol I KF) using the same template-primer used in the analysis with Pol 7. Mispair excision was examined using template-primer concentrations of 0.5 pM and 2.5 pM and AMP concentrations of 0 mM, 0.56 mM and 2.4 mM (data not shown). At the DNA concentraion of 2.5 pM, the 3'-5' exonuclease associated with Pol I KF wa§,inhibited 31% and 67% at AMP concentrations of 0.56 and 2.4 mM, respectively. The 150 (that concentration of inhibitor providing 50% inhibition) of AMP for the 3'—5' exonucleases associated with Pol I KF and Drosophila Pol Y is 1.5 mM and 3 mM, respectively, at a DNA concentration of 2.5 pM. This indicates that 5'AMP is only a 2-fold more effective inhibitor of the 3'-5' exonuclease associated with Pol I KF than the 3'-5' exonuclease associated with the Drosophila y POlymerase. Further, using a DNA concentration of 0.5 pM, the 3'-5' exonuclease associated with Pol I KF is inhibited 67% and 73% at AMP concentrations of 0.56 mM and 2.4 mM, respectively, while that associated with the Drosophila y polymerase is only inhibited 20.7% and 45.1%, respectively. Differential excision. of .3'-terminal. mispairs Several proofreading exonucleases associated with DNA P°1Ymerases from both prokaryotic and eukaryotic sources exhibit differential mispair excision, resulting in unique hierarchies of editing efficiencies (162,169,173,213). To examine the SpecifiCity’of mispair excision by the Drosophila y polymerase, a Series of natural DNA template-primers was constructed to generate dAMP:dAMP, dGMP:dGMP, and dGMP:dAMP 3'-termini 112 following annealing to single-stranded phage M13 DNA template. Time course analysis revealed no preferential excision with regard to the mispair presented (Fig. 16). Thus the efficiency of exonucleolytic editing of the specific mispairs examined on a natural DNA template-primer are similar. This is a result consistent with the proofreading exonuclease associated with the chick embryo y polymerase (162). Specificity of mispair excision DNA polymerase 7 was assayed, in the same reaction, on phage M13 DNA templates to which were annealed either a 15 nucleotide primer to generate a 3'terminal dGMP:dGMP mispair, or a 17 nucleotide primer to generate a 3'-terminal dGMP:dCMP base pair. Under nonpolymerization conditions with equimolar concentrations of 3'-terminal base pairs and mispairs, DNA polymerase 7 catalyzed hydrolysis of the dGMP:dGMP mispair at a rate ~15-fold greater rate that of the dGMP:dCMP base pair (Fig. 17A). Further, examination of the 3'-5' exonuclease assay products clearly reveals the mispair specificity of the 3'-5' exonuclease (Fig. 178). A 16-mer product, derived from the excision of a correctly paired 3'-terminus of the 17-mer primer, is not visible until 20 minutes of incubation, while linearity of the mispair excision reaction is lost after ~4 minutes of incubation (Fig 17A). 113 Figure 16. Time course of hydrolysis of 3'-terminal mispairs. DNA polymerase 7 Fr. VI (0.21 units) was assayed for 3'-5' exonuclease activity using singly-primed M13 DNA (4 pM) containing 3'-terminal dGMP:dGMP (open squares), dGMP:dAMP (open circles) or dAMP:dAMP (closed circles) mispairs. 114 100 90‘ a . ' P o 4. n o s 80 " 70 60 1' avewhdouohm .uaemewt 25 20 15 10 minutes Time, Figure 16 115 Figure 17. Examination of ndspair specificity of the 3'- 5' exonuclease associated with y polymerase. A. DNA polymerase 7 Fr. VI (0.15 units) was assayed for 3'-5' exonuclease activity using singly-primed M13 DNA (4 pM) containing a 3'-terminal dGMP:dGMP mispair(open circles) or a 3'-terminal dGMP:dCMP base pair (closed circles). Terminal mispair excision was analyzed as described in "Methods" at the time intervals of 0,2,4,10,20,30, and 40 minutes. 3. Terminal mispair or base pair excision was analyzed by gel electrophoresis as described under "Methods." Lane 1 represents a no enzyme control, lanes 2-8 correspond to the time intervals indicated above. % 3'—terminal excison, 100 90 80 70 60 50 40 30 20 10 116 O .0 0 0 e 3 ... l l l l l 1 l 10 15 20 25 30 35 40 Time, minutes 1 i3 4 5 6 7 8 117 Exonucleolytic editing during DNA synthesis and next nucleotide effect To determine the effect of an over abundance of the next correct nucleotide required for DNA synthesis prior to mispair excision, a primer extension time course analysis was performed. Template-primers possessing dGMP:dGMP and dAMP:dTMP 3'-termini (see "Methods") were utilized under DNA polymerization conditions in the presence of 30 pM or 1 mM dTTP (the next correct nucleotide required for primer extension prior to mispair excision), and in the absence of dCTP (the nucleotide required for primer extension following mispair excision). The basic premise of this analysis was to force the Drosophila mtDNA polymerase to extend a 3'-terminal dGMP:dGMP or dGMP:dAMP mispair. ‘ The data in Figure 18 indicate that in the presence of a large next nucleotide bias, extension of the dGMP:dGMP mispair to an expected 23-mer does not occur. Instead, quantitative mispair excision is observed over the time course of incubation. In addition, the rate of mispair excision is nearly equal in the presence and absence of dNTPs (Fig. 16 and Fig. 20), indicating that it is not affected by the presence of dNTPs. The same experiment was performed using a template-primer containing a 3'-terminal dAMP:dTMP base pair. The appearance of an expected 21-mer and an initially unexpected 24-mer correspond to the rate of disappearance of the 15-mer. This indicates that 1 mM dTTP is not inhibitory to primer extension (Fig. 19). 118 Figure 18. Template-primer utilization by Drosophila DNA polymerase y -- next nucleotide effect under ”stationary conditions." DNA polymerase y Fr. VI (0.17 units) was incubated with 5'end-labeled singly-primed M13 DNA (4 pM) containing a 3'-terminal dGMP:dGMP mispair under standard DNA polymerase assay conditions as described in "Methods,” in the absence of dCTP and the presence of 30 pM dTTP (A) or 1 mM dTTP (3). Product analysis was performed by denaturing gel electrophoresis and autoradiography of reactions incubated for 0,2,4,10,20,30,and 40 min. Lanes 1 and 9 represent the no enzyme controls, lanes 2-8 and 10-16 correspond to the time intervals indicated above. 12 3 456 7 8 910111213141516 —-— lS—mer ——— 14-mer Figure 18 120 Figure 19. Template-primer utilization by Drosophila DNA polymerase y -- next nucleotide effect under ”synthesis conditions.” DNA polymerase y Fr. VI (0.17 units) was incubated with M13 DNA (4 pM) containing a 3'-terminal dAMP:dTMP paired primer under standard DNA polymerase assay conditions in the absence of dCTP and in the presence of 30 pM (A) or 1 mM TTP (B). Product analysis was as described in Figure 6. 12345678 910111213141516 .3 -6, '.' . ..‘_.. -. '\ ‘.' '4.‘ ‘ (A _ ‘ , _ . .. : _ . , 4" 0.,“ ‘I . -. A . ' ~ . ° ~§ ' .. ‘. ' _ . ‘n .' . ~.. 1.: ‘ ..' I "-“,- .vo. '- ' . ..‘ '0‘. r : -.‘_ I‘ g ' '4 I". . - . ve‘ . ". ' ‘ ’> ‘ I I . e... ‘ ‘ 1 I l ‘ -—- lS—mer -___... ...-.._--._1_. Figure 19 122 Upon examination of the template DNA strand (see inset to Figure 19) it is clear that the formation of the 24-mer results either from the incorporation of contaminating dCTP or of an incorreCt nucleotide at position 22. The abundance of the 24-mer is 3- fold greater in the presence of 1 mM dTTP. Lack of primer extension from the dGMP:dGMP 3—OH terminus following mispair excision indicates no dCTP contamination (Fig. 18) and previous results showed that in the presence of dCTP, this primer will be extended following mispair excision (163). Taken together, the data indicate that when in a "synthesis" mode, the Drosophila y polymerase will misinsert and then extend a misincorporated nucleotide opposite a template dGMP. Notably, the rate of misincorporation and subsequent primer extension is > 40-fold lower than the rate of primer extension from a correctly paired primer in the presence of 30 pM TTP, and 15-fold lower in the presence of 1 mM dTTP. This was determined by comparing the rate of primer extension from the 15-mer to the 21-mer (no misinsertion) to that from the 21-mer to the 24-mer (misinsertion at position 22). This experiment also suggests that misinsertion and not primer extension of a mispair is rate limiting; if the 15-mer were extended to a 22-mer, then primer extension of a mispair would be rate limiting. Analysis of Primer extension product lengths reveals pausing occurs at Positions 21 and 24 on the template, prior to misinsertion OPposite a template dGMP residue. 123 Figure 20. Quantitation of template-primer utilization by Drosophila y polymerase The data in Figures 6 and 7 were quantitated by densitometric scanning. Open and closed squares represent the rate of primer extension of the dAMP:dTMP base pair in the absence of dCTP and in the presence of 30 pM and 1 mM TTP, respectively. Open and closed circles represent the rate of hydrolysis of the (dGMP:dGMP) mispair in the absence of dCTP and in the presence of 30 pM and 1 mM TTP, respectively. 124 100 w 63339 ounuuunsm 15 20 25 minutes 10 Time, Figure 20 125 Mechanism of template-primer transfer from 3‘-5' exonuclease to 5'-3' DNA polymerase site Previous analysis of the mechanism of proofreading by the Drosophila Y polymerase indicated that mispair hydrolysis was the rate limiting step (163). No accumulation of mispair excision products following the necessary hydrolytic step was observed. Two mechanisms of template-primer transfer from the 3'-5' exonuclease site to the DNA polymerase site are possible. The first involves rapid dissociation and reassociation of the same or another 7 polymerase molecule following mispair excision. The second involves template-primer transfer from the 3'-5' exonuclease site to the DNA polymerase site in which the enzyme remains associated with the DNA. To distinguish between intermolecular and intramolecular shuttling of the template- primer from the 3'-5' exonuclease site to the DNA polymerase site of Pol % a strategy was devised to initiate the mispair excision reaction and then allow for primer extension of the resulting correctly paired 3'—termini in the presence of a large excess of unlabeled DNA (a DNA trap). If the template-primer is passed from the 3'-5' exonuclease site to the DNA polymerase site via an intermolecular mechanism, then primer extension products (normally formed in the absence of the DNA trap) would not be observed. Conversely, if transfer of the labeled template-primer was intramolecular, then primer extension products from the correctly paired 3'—termini should be observed because the DNA polymerase remains bound to the template-primer. That is, one round of processive DNA synthesis by the Y 126 polymerase should be observed on the singly-primed DNA substrate following mispair excision, unaffected by the presence of the DNA trap. The results of such an experiment using the M13 exonuclease substrate containing a dGMP:dGMP mispair (see "Methods") are shown in Figure 21. In the absence of the DNA trap and dNTPs, Pol y removes ~50% of the 3'-terminal mispairs within 20 minutes (lane 2). In the absence of the DNA trap but in the presence of dCTP, dGTP and dTTP Pol 7 extends within 20 minutes ~50% the 15 nucleotide primer to the expected l9-mer following 3'-terminal mispair excision (lane 4). This control reveals that 7 polymerase prefers to extend paired 3'-termini as opposed to excising mispaired 3'-termini. In addition, primer extension appears to occur immediately following mispair excision, since no products of mispair excision (14-mers) are observed. The lack of mispair excision and of primer extension following mispair excision in the presence of the DNA trap (lanes 3 and 5) suggests the DNA trap effectively quenches both reactions. A time course of primer extension in the presence of the DNA trap following ~ 50% removal of the 3'-terminal mispairs (lanes 6 to 12) revealed less than 8% of the available 14 nucleotide primers were extended to a 19-mer after 40 minutes of incubation. At the same time, a control experiment performed under the same reaction conditions but with a template-primer with a correctly paired 3'-terminus (dGMP:dCMP l7-mer, see "Methods") revealed 97% primer extension after 10 minutes of 127 Figure 21. Transfer of the template-primer from the 3'- 5' exonuclease to the DNA polymerase site of Drosophila Pol y. DNA polymerase y (Fr. VI, 0.2 units) was incubated with the M13 exonuclease substrate (dGMP:dGMP mispair) as described in "Methods" with the exceptions indicated below. Primer extension in the presence of the DNA trap, dCTP, dGTP, and dTTP (following ~50% mispair excision) was analyzed at the time intervals of 0, 2, 4, 10, 20, 30 and 40 minutes. Lane 1 represents a no enzyme control. Lane 2 represents a 20 minute incubation of the 3'-terminal mispaired exonuclease substrate in the absence of dNTPs and the DNA trap. Lane 3 is a 20 minute incubation as in lane 2 but in the presence of the DNA trap. Lane 4 is as in lane 2 except in the presence of dCTP, dGTP, and dTTP. Lane 5 is as in lane 4 except that the DNA trap was included. Lanes 6-12 correspond to the time intervals indicated above. 1 2 3 4 5 6 7 8 9 10 11 12 - — 1 9—1ner Figure 21 129 I incubation in the absence of the DNA trap (data not shown). : contrast, only 2% of the available l4-mers were extended at the 10 minute time point in the presence of the DNA trap. Limited primer extension from both the 14-mer (19-mer product, Fig.21) and the 17-mer (21—mer product, data not shown) in the presence of the DNA trap indicates that the DNA trap is ~90% efficient. The data obtained from the control primer extension (data not shown) indicates that the limited DNA synthesis on the M13 DNA substrate in Figure 21 most likely results from a limitation in the DNA trap and is not due to intramolecular transfer of the template-primer from the 3'-5' exonuclease site to the DNA polymerase site. Furthermore, if the y polymerase were to remain bound to the template-primer, and transfer of the DNA substrate to the DNA polymerase site occured via an intramolecular mechanism, then the rate of primer extension of the 14-mer to the 19-mer should be comparable to the rate of primer extension in the absence of the DNA trap of a correctly paired 3'-terminus in a single round of processive DNA synthesis. Instead, the rate of primer extension of the 14—mer in the presence of the DNA trap is >50-fold lower than the rate of primer extension in the absence of the DNA trap. The lack of primer extension of the l4-mer, produced following excision of the 3-terminal mispair from the 15-mer, in the presence of the DNA trap indicates that Y polymerase does not remain associated with the template-primer following mispair excision. Instead, the data suggest a mechanism where DNA synthesis following mispair excision requires the rapid 130 reassociation of the same or second enzyme molecule to the correctly paired 3'-terminus. Therefore, template-primer transfer from the 3'-5' exonuclease site to the DNA polymerase site apparently requires dissociation of the y polymerase prior to initiation of DNA synthesis. 130 DISCUSSION Because of the apparent lack of a DNA repair mechanism in mitochondria, the association of a proofreading exonuclease with the mtDNA polymerase could be essential for the maintenance of genetic integrity of the mitochondrial genome. Previous studies have demonstrated that a proofreading exonuclease can increase the fidelity of DNA replication up to 100-fold (120,138,219). However, a study performed with E. coli DNA polymerase I Klenow fragment and genetically engineered Exo‘ mutants indicates that the greatest factor influencing the accuracy of DNA replication is discrimination in nucleotide selection by the DNA polymerase function (115). The most fundamental and probably the most important characteristic of a proofreading exonuclease is its ability to function coordinately and efficiently with the DNA polymerase. To demonstrate the ability of the 3'-5' exonuclease in Drosophila Pol y to function coordinately with the DNA polymerase, the KCl optima were determined for both activities on two templates. The data demonstrate that the 3'—5' exonuclease functions optimally in the same or similar ionic environment as the DNA polymerase, using two template-primers differing significantly in primer density. A survey of the monovalent cation requirements of DNA polymerases and their corresponding proofreading exonucleases from several sources, suggests that they are similar if not identical (76,138,150,161,162,171,221,228,229,230,231,232, 233). The DNA polymerase and 3'-5’ exonuclease activities associated with the chick embryo Y polymerase are assayed in the presence of 110 and 131 150 mM KCl, respectively (162). The porcine liver mtDNA polymerase and its associated 3'-5' exonuclease apparently require 150 mM and 120 mM KCl, respectively for optimal activity (161). On the otherhand, the Xenopus Y polymerase apparently requires a K01 concentration of 100 mM for the DNA polymerase activity (160; while its the 3'-5' exonuclease is assayed at a KCl concentration of 50 mM (164). The cryptic and potent mismatch-specific proofreading exonuclease associated with the Drosophila a polymerase was detected under the same reaction conditions (in the absence of dNTPs) as the DNA polymerase activity (138). Likewise, both the DNA polymerase and 3'-5' exonuclease activities of 5 polymerase as purified from human placenta and CV-l cells are KCl sensitive (150,230). However, the 5 polymerase isolated from calf thymus exhibits KCl sensitivity only with regard to the DNA polymerase activity (231); its 3'-5' exonuclease activity, though active under low salt conditions, exhibits optimal activity at 80 to 100 mM KCl. The divalent metal cation (Mg2+) optimum of Drosophila y polymerase is nearly identical for the DNA polymerase and 3'-5' exonuclease activities. Similarly, the 3'-5' exonuclease activities associated with other DNA polymerases, isolated from both prokaryotic and eukaryotic sources, apparently require the same MgC12 concentrations as their respective DNA polymerase activities (138,161,162,171,213,233,234). The template—primer affinity of the 3'-5' exonuclease and DNA polymerase activities of Drosophila 7 polymerase varies less than 132 1.4-fold on singly-primed M13 DNA. Similarly, upon association :5 the E. coli DNA polymerase III a- and e-subunits, which contain the polymerase and exonuclease catalytic activities, respectively, the template-primer affinity for both activities varies less than 2— fold (171). The DNA polymerase and 3'—5' exonuclease activities of Drosophila Pol 7 have been examined on several DNA template- primers. In doing so, the DNA polymerase to 3'-5' exonuclease ratios have been determined and shown to vary between 1700:1 on singly-primed M13 DNA and 2.5:1 on poly dA - oligo dT15. This ratio for E. coli Pol I Kf varies similarly between 8000:1 on singly-primed M13 DNA and 2:1 on poly dA - oligo dT15. A hallmark of proofreading exonucleases has been their selective inhibition, presumably by end-product inhibition, by nucleoside monophosphates (NMPs) (76,143,151,161,167,204,218,224, 225). Probably the most comprehensive study with regard to NMP inhibition of proofreading exonucleases was performed with Pol I Kf (218). Here, the Ki for dTMP was determined to be 0.13 mM and Lineweaver-Burk analysis demonstrated it to be a competitive inhibitor of the DNA substrate poly deo. The Ki for AMP for the 3'-5' exonuclease associated with Drosophila y polymerase was ~3 mM when analyzed as a competitive inhibitor relative to 3'-OH mispaired termini on a partially duplex substrate: the data in showed an increased inhibition of 3'-5' exonuclease as a function of AMP concentration and as an inverse function of DNA concentration, consistent with a mode of competitive inhibition. However, the high Ki relative to the Km of 133 the DNA substrate (1.38 pM) indicates the relative inefficiency :5 this hydrolytic product as a competitive inhibitor of mismatched 3'-termini for the proofreading exonuclease. A comparative analysis indicated that AMP inhibits exonucleolytic excision of 3'-teminal mispairs by Pol I KF only 3-fold more effectively than that by Pol M The relatively low inhibition of the 3'-5' exonuclease of Drosophila Pol 7 by NMPs is not inconsistant with data presented in previous studies (161,162,170,213,225,23l,234). Although both direct inhibition of 3'-5' exonuclease activity and indirect measurements of decreased fidelity of DNA synthesis, have been described, NMPs must be present at concentrations three to four orders of magnitude greater than the concentrations 3'-terminal mispairs to effectively inhibit the 3’-5' exonucleases. Examination of the chick embryo y polymerase revealed a 54 and 69% inhibition by 10 mM AMP and GMP, respectively (162). The porcine liver mtDNA polymerase exhibited a 19% reduction in exonuclease function in the presence of 20 mM GMP under DNA synthesis conditions (161). The same can also be said for the 3'-5' exonuclease associated with 8 polymerase from calf-thymus (151). Here, 5 and 10 mM AMP inhibited 3'-5' exonuclease activity only 20% and 25%, respectively, under DNA synthesis conditions. On the other hand, the proofreading exonuclease of Drosophila Pol a.was inhibited 3- and 6-fold on 3'-terminal dAMP:dAMP and a dAMP:dCMP mispaired substrates, respectively, in the presence of 10 mM GMP (173). However, the fidelity of DNA synthesis, increased under these 134 conditions as demonstrated by a 4-fold reduction in the revers;:n frequency of the ¢Xl74am3 mutation. This indicates that the effect of NMPs is not always to decrease the replication fidelit" Thus while NMPs may inhibit exonuclease activity under nonpolymerization conditions, the observed increase in the fidelity of DNA synthesis by Pol a in the presence of NMPs, suggests a lack of inhibition of the 3'-5"exonuclease under conditions in which proofreading DNA synthesis should occur. Drosophila Y polymerase hydrolyzes a 3'-terminal dGMP:dGMP mispair ~15-fold more efficiently than a correctly paired dGMP:dCMP 3'-terminus. Comparable results were previously obtained when a synthetic template-primers with 3'-terminal dAMP:dAMP or dAMP:dCMP mispairs were examined (163). In that study, the 3'~5' exonuclease was shown to hydrolyze the dAMP:dCMP and dAMP:dAMP mispairs 14- to 34-fold more efficiently than a correctly paired dAMP:dTMP 3'-terminus. Similarly, the porcine liver 7 polymerase excises mispaired 3'-termini ~7-fold more efficiently than correctly paired 3'-termini when a singly-primed natural DNA template is used (161). The chick embryo and Xenopus Y polymerases have been shown to excise preferentially mispaired over paired 3'-termini. However, a series of products corresponding to the hydrolysis of correctly paired 3'-termini, both following the excision of the mispaired, and by initially correctly paired 3’-termini is also generated (161,162). The proofreading exonucleases of E. coli DNA polymerase I and DNA polymerase III core, DNA polymerase a and DNA polymerase 5 135 (as isolated from various sources) have also been shown to excise preferentially mispaired 3'-termini (138,150,17l,l73,l74, 190,221). Notably, DNA polymerase III core and the isolated 8- subunit hydrolyze mispaired termini 10— and 45-fold more efficiently, respectively, than correctly paired 3'-termini (17l). Further, the 3'-5' exonuclease associated with Pol I Kf also appears mispair specific when a singly—primed M13 DNA substrate is used (115). However, both T5 DNA polymerase and E. coli DNA polymerase III HE, in the presence of ATP, will processively hydrolyze a correctly base paired primer in a 3'-5' direction, in the absence of dNTPs (222,235). Differential hydrolysis of several purinezpurine mispairs by Drosophila mtDNA polymerase was not observed. This is in contrast to differential mispair excision demonstrated for T4 DNA polymerase and Drosophila a polymerase using a similar DNA substrate (173,213). T4 DNA polymerase excises dAMP:dAMP mispairs at a greater rate than dGMP:dGMP and dGMP:dAMP mispairs (173). Drosophila a polymerase excises dAMP:dCMP and dAMP:dAMP mispairs 2.6- and 1.45-fold more effectively than a correctly paired dAMP:dTMP 3'-terminus. Notably, a dAMP:dGMP mispair is hydrolyzed 3.5-fold less efficiently than a 3'-terminal dAMP:dTMP base pair (213). Previously, Drosophila Pol 7 had been shown to be incapable of either extending a mispair or misincorporating a nucleotide from a "stationary" state (163). However, this enzyme exhibits m increeased misincorporation of nucleotides in the presence of nucleotide pool bias (167). Therefore, we wished to discern if 136 Pol 7 would misinsert a nucleotide and/or extend a 3'-terminal mispair from a "stationary" state, under conditions of a large next nucleotide bias. The results demonstrate that the Drosophila y polymerase is neither capable of misinserting, nor extending a 3'-terminal mispair when in a "stationary" state even in the presence of a large excess of the next nucleotide. The same is apparently true for the porcine liver and the chick embryo y polymerases (161,162). Both enzymes excise a 3'-terminal mispair with an efficiency approaching 99% prior to primer extension. Furthermore, the efficiency of 3'-terminal mispair excision by the porcine liver 7 polymerase is only slightly reduced (to 95%) in the presence of dNTPs at a concentration of 2 mM as opposed to 50 pM (161). Primer extension analysis in the absence of dCTP and using a primer with a correctly paired 3'—terminus revealed that the Drosophila y polymerase will misincorporate a nucleotide and extend a mispaired 3'- terminus while in a "synthesis state." At the same time, lack of DNA synthesis from a dGMP:dGMP mismatched primer indicates the absence of dCTP contamination, suggesting that the primer extension observed is due to a misincorporation event. Analysis of the template sequence and given the sizes of the primer extension products, which accumulate at a position prior to misincorporation opposite a template dGMP residue, indicates that the postulated misincorporation event, and not primer extension of the mispair, is rate limiting. If the 15 nucleotide primer was shown to be extended to a 22 nucleotide product (indicating pausing following misincorporation), then 137 presumably the proofreading exonuclease would have an opportunity to excise the mispair. This suggests 21-mer and 22-mer products would be detected by electrophoretic analysis over the time course used. In addition, formation of the 24-mer product suggests some proportion of the Pol 7 molecules must remain associated with the template-primer following the misincorporation event; if each mtDNA polymerase dissociated following the misincorporation event, then presumably the y polymerase would bind to the template-primer in a "stationary" state providing the opportunity for the 3'-5' exonuclease to excise the newly formed mispair. Nucleotide misinsertion and the extension of a mispair by Drosophila y polymerase in a "synthesis" state is perhaps not unexpected in the reaction carried out in the absence of a required nucleotide or under nucleotide pool bias conditions. Fidelity experiments using the duu74 am3 or the Lacza forward mutational assays have demonstrated DNA polymerases will misinsert a nucleotide and extend a mispair (138,l42,151,161,162,l67,170, 213,236). In particular, work by Mendelmann et al. revealed that Drosophila a polymerase and AMV reverse transcriptase, both devoid of detectable 3'-5' exonuclease activity, will misincorporate nucleotides when either in a "synthesis" or "stationary" state (236). However, the Km for inserting the wrong nucleotide was two to three orders of magnitude greater than that for insertion of the correct nucleotide when either in a "synthesis" or "stationary" state. In addition, a polymerase was shown to pause prior to nucleotide misincorporation, suggesting this, rather than 138 elongation from a 3'-terminal mispair, is the rate limiting Step (236). Two mechanisms of template-primer transfer from the 3'-5' exonuclease site to the DNA polymerase site of Drosophila y polymerase are possible. The first is transfer of the template- primer to the DNA polymerase site from the 3'-5' exonuclease site without dissociation of the enzyme-DNA complex. The second is one in which the template-primer at the 3'—5' exonuclease site is dissociated following the mispair excision reaction and rebinding to the DNA polymerase active site of Pol y for subsequent DNA synthesis. The data in this report demonstrate that following mispair excision, a newly formed paired primer is not extended in the presence of challenger DNA, indicating that the Drosophila y polymerase must dissociate from the template-primer following mispair excision. Exonucleolytic hydrolysis was allowed to proceed during a preincubation stage prior to the addition of the challenger DNA. Thus, following mispair excision, movement of the template-primer from to the DNA polymerase site via and intermolecular mechanism was not prevented. Nevertheless, primer extension products (l9-mer) derived from the paired primer (14- mer) were not observed in the presence of the DNA trap. Therefore, primer extension following the necessary excision reaction to remove the 3'-terminal mispair appears to occur after the association of a second Pol 7 molecule. This is contrary to the mechanism of template-primer transfer determined for bacteriophage T5 DNA polymerase and E. coli DNA 139 polymerase I Klenow fragment (175,222,237). Mizrahi et al. 2’ be I h 0' (n H demonstrated that following exonucleolytic excision, Pol capable of a single round of processive DNA synthesis prior to cycling to the challenger DNA. In addition, the rate of disappearance of the exonuclease excision products appeared to be greater than that of the primer molecules that had not been reduced one nucleotide in length, again suggesting intramolecular template-primer transfer. A second analysis by Joyce involved a preincubation in the absence of Mg2+ to prevent artifactual cycling of Pol I Kf (175). Following the simultaneous addition of Mg2+, challenger DNA and dNTPs, 40 to 50% of mispair excision products generated were extended, suggesting intramolecular transfer of the template- primer from the 3'—5' exonuclease active site to the DNA polymerase active site. In summary, the data presented here indicate the DNA polymerase and 3'-5' exonuclease activites with the mtDNA polymerase are functionally cooridinated, that primer extension from a 3'- terminal mispair will not occur even under conditions of a large next nucleotide bias, that nucleotide misincorporation of is rate- limiting relative to the extension of a mispair, and that the mechanism of template-primer transfer from the 3'-5' exonuclease site to the DNA polymerase site of Drosophila y polymerase is intermolecular. CHAPTER 17 Structure-Function Relationships in Drosophila 'y Polymerase INTRODUCTION The mitochondrial DNA polymerase is a nuclear-encoded enzyme which has been highly purified from four sources: chick embryos (158), Drosophila embryos (159), X. laevis oocytes (160) and porcine liver (161). As purified from each source, y polymerase exhibits sensitivity to the alkylating agent N-ethylmaleimid is inhibited by dideoxynucleoside triphosphates and is resistant to inhibition by the drug aphidicolin. Taken together, the above characteristics distinguish Pol y from the nuclear DNA polymerases a, B, 6, and 8. In addition, DNA polymerase y is distinguished from other DNA polymerases by its relatively high monovalent cation optima: the optimum is 200 mM KCl when DNase I-activated calf thymus DNA is the DNA substrate (159,160) and is greater than 100 mM when poly dA - oligo dT is the template- primer (159-161). Finally, a 3'-5' exonuclease is associated with each of these mitochondrial DNA polymerases (161-164). Gel filtration and velocity sedimentation of the purified mitochondrial DNA polymerases suggest a native molecular weight of approximately 160 to 180 kDa (158-161). The near-homogeneous chick embryo and Drosophila y polymerases have been shown to be composed of two polypeptides. SDS-polyacrylamide gel electrophoresis analysis revealed the presence of a 47 kDa polypeptide (major polypeptide) and a 135 kDa polypeptide (minor 140 141 polypeptide) for the chick embryo enzyme (158). The near- homogeneous Drosophila enzyme is a heterodimer comprising subunits of 125 kDa and 35 kDa (159). In both cases the presence of these polypeptides correlate with DNA polymerase activity (158,159). SDS-polyacrylamide gel electrophoresis of the most pure porcine liver and Xenopus Y polymerase preparations has demonstrated the presence of polypeptides of 120, 55, 50, and 48 kDa in the former (160) and 140, 100, 85, 55, 40, and 31 kDa in the latter (161). The DNA polymerase function in mitochondrial DNA polymerase was shown to be associated with the 125 kDa subunit in the Drosophila enzyme (159). Cross-linking of BrdUMP-substituted DNA to the X; laevis Y polymerase, presumably at the template— primer binding site, resulted in the ultra-violet exposure dependent labeling of the 140, 100, and 55 kDa polypeptides (160). Further, ethylene glycol-glycerol gradient sedimentation of the X. laevis Y polymerase also suggested the 140 kDa polypeptide is the catalytic subunit (164). Taken together, the data from native molecular weight determination, subunit composition and polypeptide activity analyses suggest the mitochondrial DNA polymerase is a heterodimer containing a large catalytic polypeptide of 125 to 140 kDa and 35 to 48 kDa polypeptide of undefined function. That the smaller polypeptide is not a proteolytic product of the larger one has not been determined. It might be generated during enzyme purification as observed with early purifications of DNA polymerases 0: and 5 (see 137 for review). 142 To demonstrate that the smaller polypeptide, the 35 kDa or B-subunit, is a distinct component of the Drosophila mitochondrial DNA polymerase, several approaches could be employed. First, dissociation and separation of the two subunits with the retention of catalytic activity might allow the subunit assignment of the 3'-5' exonuclease. In addition, the isolated 125 kDa (catalytic subunit) could be re-analyzed with regard to the kinetic parameters for DNA polymerase activity and compared to those established characteristics associated with the Y polymerase holoenzyme. This type of analysis has the potential of providing insight as to the contribution of the B-subunit, with regard to the isolated DNA polymerase and 3'-5' exonuclease activities and in terms of (holoenzyme function. Subunit dissociation and separation has been performed successfully with the Drosophila a polymerase by glycerol gradient sedimentation in the presence of urea and ethylene glycol (138,191). These studies revealed that the DNA primase activity is associated with the 50 and 60 kDa polypeptides. Further, the DNA polymerase and a cryptic 3'-5' proofreading exonuclease were shown to be associated with the 180 kDa subunit. Glycerol gradient sedimentation in the presence of 50% ethylene glycol was also performed with the X. laevis Y polymerase (164). However, the larger 140 kDa polypeptide was not separated from several smaller polypeptides of approximately 40 kDa. 143 A second approach to evaluating the roles of the two subunits of Drosophila Pol Y involves the development of potent and specific polyclonal antiserum to near-homogeneous enzyme. Immunoprecipitation of rapidly prepared embryo and mitochondrial homogenates and/or crude enzyme fractions could provide information regarding the subunit structure of the enzyme prior to substantial purification. Further, proteolytic cleavage of the separated a- and B-subunits and subsequent SDS- polyacrylamide gel electrophoresis of the proteolyzed polypeptides would indicate their structural relatedness. 144 RESULTS Specificity' of“ antiserum. developed. against the Drosophila Y polymerase. An immunoblot analysis was used to determine which subunits of the Y polymerase were recognized by antiserum derived by rabbit immunizations (see "Methods"). DNA polymerase Y of ~8% purity was subjected to immunoblotting as described in "Methods" and the legend to Figure 22. The data in Figure 22 indicate that the antiserum developed againsg’near- homogeneous Y polymerase (Fraction VI) specifically recognizes two polypeptides of 125 and 35 kDa in molecular weight. These presumably correspond to the a- and B-subunits of the DroSOphila mtDNA polymerase, respectively. The same enzyme fraction was subjected to immunoprecipitation using the anti-Pol Y immune and pre-immune rabbit sera, as described in ”Methods" and the legend to Figure 23. Here, the detection of a 125 and a 35 kDa polypeptide (and the IgG heavy-chain) only from the immune serum-containing sample is observed upon subsequent immunoblotting. The data indicates that neither the or nor the B-subunits have been proteolyzed during the purification procedure from the Fr.III to the Fr.VI stage (159). Inhibition of DNA polymerase and 3'-5' exonuclease by anti-DNA polymerase Y serum. To demonstrate that the antiserum is highly specific for the Drosophila Y polymerase, inhibition studies of the DNA polymerase and the 3'-5' exonuclease were performed. Antiserum to the DrOSOphila Y polymerase was incubated with DNA polymerase Y as described in "Methods" at the concentrations indicated in Figure 24. The Figure 22. Reactivity of rabbit antiserum against DNA polymerase y. Pol 7 Fraction IV (159) 12 units (lane 1, 1.9 pg total, ~10% pure) or 24 units (lane 2, 3.9pg total) was denatured, electrophoresed in a 10% SDS polyacrylamide gel and transferred to nitrocellulose. The blot was incubated with the rabbit antiserum (1:1000) for 2 hours at 24°C and then probed with [125I1-protein A as described in ”Methods". This data was provided by Dr. Rhoderick H. Elder. B— 146 Figure 22 147 Figure 23. Immunoprecipitation of DNA polymerase y. DNA polymerase Y Fraction III (19.7 units, 3.4 pg total protein, ~8% pure) was incubated overnight with Pol Y antiserum (lane 1) or rabbit pre-immune serum (lane 2) (0.67% in 10 mM NaP04, pH 7.0, and 154 mM NaCl). Immune complexes were precipitated by incubation with 1/6 volume of pre-swollen protein A—agarose in 0.67% in 10 mM NaPoq, pH 7.0 and 154 mM NaCl and then centrifuged. The immune complexes were washed, suspended in 1x Laemmli sample buffer, and subjected to immunoblot analysis as described in legend to Figure 22. This data was provided by Dr. R. H. Elder. I48 Fritz?" Od—a ,— IgG H—chain Figure 23 149 Figure 24. Inhibition of DNA polymerase Y by rabbit antiserum. (A) DNA polymerase Y Fraction VI (0.3 units) was preincubated for 1 hour at 0°C with the indicated concentrations of pre-immune (closed circles) or immune (open circles) serum. (Levels are expressed as a percentage of the final reaction volume). The samples were assayed for DNA polymerase activity under standard reaction conditions on singly-primed M13 DNA as described in ”Methods.” (3) Inhibition of 3'-5' exonuclease activity. Pol Y Fraction VI (0.25 units) was preincubated as described above and then assayed for 3'-5' exonuclease activity using singly-primed M13 DNA under standard reaction conditions as described in "Methods." The data in Panel A was provided by Dr. R.H. Elder. 150 L fix . o o o m 9 8 a .ocasaeEem aua>auo¢. eeeuefihaom .azo \. n o 7 b o 6 - o 5 n o T D b o 3 I 20 ‘ L — o 1.. b .20 .15 .10 .05 0.00 % Serumn Concentrations . D b b n n n p o o o o 4. 3 2 1 e.uddsdeaom hud>duo¢ ensedoasowm 1 1 0.05 .10 0.15 0.20 00 Serum Concentration,% Figure 24 151 data presented in Figure 24A and 24B demonstrate that the rabbit antiserum inhibits both the DNA polymerase and the 3'-5' exonuclease activities. The preimmune serum has no significant effect on either activity. The 3'-5’ exonuclease and DNA polymerase as assayed on singly-primed M13 DNA (containing a 3'— terminal mispair for the exonuclease assay) are inhibited to nearly the same extent at each rabbit antiserum concentration tested. For example, at an antiserum concentration of 0.2% the 3'—5' exonuclease and DNA polymerase are inhibited 70% and 60%, respectively. However, higher concentrations did not result in complete inhibition of the 3'-5' exonuclease and DNA polymerase associated with the Y polymerase. This is most likely due to exclusion of activity neutralizing antibodies by the binding of other non-neutralizing antibodies or the binding of antibodies to one active site blocks the binding of antibodies to a second active site. Under conditions where the antiserum produced maximal inhibition of the Drosophila Y polymerase, no inhibition of the DNA polymerase or 3'-5' exonuclease activities associated with E. coli DNA polymerase I or its Klenow fragment were observed (data not shown). Dissociation and separation of the subunits of Drosophila Y polymerase The Drosophila mtDNA polymerase consists of a 125 kDa DNA polymerase catalytic subunit and a 35 kDa subunit of unknown function (159). To elucidate the structure-function relationships in the mtDNA polymerase and to make a subunit assignment for the 3'-5' exonuclease associated 152 with Pol Y, extensive studies to achieve dissociation and separation of the its two subunits with retention of catalytic activity were pursued. Glycerol gradient sedimentation of the Drosophila Y polymerase in the presence of 50% ethylene glycol was performed as described under "Methods." The precedent for this procedure was the successful attempt dissociation and separation of DNA polymerase a subunits(138). A similar experiments with the Drosophila Y polymerase is shown in Figure 25. E. coli DNA polymerase I (Mr 109 kDa, 5.5 S) and bovine carbonic anhydrase (BCA, Mr 30 kDa, 3.2 S) were sedimented in parallel gradient to approximate the locations of the dissociated and separated Pol Y a— and B—subunits. The results demonstrate that the DNA polymerase and the 3'-5' exonuclease activities associated with the Y polymerase cosediment in the gradient with an 86% and a 295% recovery of the two activities, respectively. However, in this experiment immunoblot analysis failed to detect either the a- or B-subunits of Pol Y. Subsequent experiments in which the a- and B-subunits were shown not to be dissociated also demonstrated near complete recovery of the DNA polymerase and 3'-5' exonuclease activities. Lack of subunit dissociation and separation by glycerol gradient sedimentation in the presence of 50% ethylene glycol is apparently responsible for the high recovery of DNA polymerase and 3'—5' exonuclease activities. 153 Figure 25. Cosedimentation of DNA polymerase and 3'-5' exonuclease in the presence of ethylene glycol. Glycerol gradient sedimentation of DNA polymerase Y in the presence of 50% ethylene glycol was performed as described in ”Methods." DNA polymerase (open circles) and 3'-5' exonuclease (closed circles) were assayed with DNase I-activated calf thymus DNA and singly-primed M13 DNA containing a 3'-terminal dAMP:dAMP mispair, respectively. Fraction 1 represents the bottom of the , gradient. E. coli DNA polymerase I (25 units, 5.5 S) and bovine carbonic anhydrase (BCA, 700 pg, 3.2 S) were sedimented in a parallel gradient and assayed for DNA polymerase and protein concentration, respectively, as described in "Methods." 5 w .cowmwoxw wwemmwz 0 5 o 2 1 1 5 0 _ . . . t .0 6 e *0 65 0‘ 8” W m- w v e the A YPE .1 l x . I“ 10 e . . . A l. l , .\.. lo .. u 3 ‘ .\ I Il.‘lll‘l 1; .. OIII- e -/.I :0 2 1.. .1.nlv o/ e - e d d u e I 1 u e u - 1‘10 8 7 6 S 4. 3 2 1 o MIOOH on Emu Sudan—Bod ensues—endow 5.8 Number Fraction Figure 25 155 Preparative scale Sephacryl S-200 gel filtration chromatography in the presence of 0.4 to 1.5 M guanidine-HCl (Gu-HCl) was performed as described under "Methods" in a further effort to dissociate and separate the Pol Y subunits. Numerous control experiments were first performed to examine the effeCts of Gu-HCl on enzyme activity. Gu-HCl‘is a chaotropic agent which, under controlled conditions, has been shown to partially unfold or denature enzymes without rendering them completely inactive (238). A representative gel filtration analysis is presented in Figure 26. Gel filtration in the presence of 0.4 Gu-HCl (Fig.26A) results in 2 95% retention of polymerase and 3'-5' exonuclease activities, but subunit separation does not occur. Immunoblot analysis demonstrates the a- and B-subunits are coincident with the two activities. E. coli DNA polymerase I and bovine carbonic anhydrase were subjected to Sephacryl S- 200 gel filtration chromatography under the same conditions to approximate the locations of the dissociated and separated a- and B-subunits. The presence of l M Gu-HCl (Fig.26B) abolished the 3'-5' exonuclease activity. However, the enzyme chromatographs at or near its native molecular weight and the a— and B-subunits are coincident with DNA polymerase activity. The loss of 3'-5' exonuclease activity suggests partial unfolding of the enzyme has occurred. Subunit separation appears to have been achieved in the presence of 1.5 M Gu-HCl (Fig. 26C), but no enzyme activity was recovered in the location of the d—subunit, which was recovered 156 Figure 26. Gel filtration of Drosophila DNA polymerase Y in guanidine hydrochloride. (A) Pol Y Fraction VI (66 units, 825 ng) was preincubated for 1 hour at 0°C in 0.4 M Gu- HCl and then chromatographed on a Sephacryl S-200 column (10 ml) as described in "Methods." Fractions (0.1 ml) were collected and aliquots (5 pl) were assayed for DNA polymerase activity on DNase I-activated calf thymus DNA and for 3'-5' exonuclease activity on singly-primed M13 DNA as described in "Methods." The a- and 8-subunits were detected by immunoblot analysis of aliquots (0.08 ml) of the column fractions following precipitation in trichlorolacetic acid and SDS-PAGE. The column was calibrated with E. coli DNA polymerase I (Pol I, 109 kDa) and bovine carbonic anhydrase (BCA, 30 kDa) which were preincubated in 0.4 M Gu-HCl as described above. (8) As in “A” except that the preincubation and chromatography buffers contained 1.0 M Gu-HCl. (C) As in ”A” except that the preincubation and chromatography buffers contained 1.5 M Gu-HCl. DNA Polymerase Actlvlty, CPM :1 109-3 157 26': 0.4M Gu-HCI [---—] Pol I Fraction Number Figure 26A YIeId Pol: >95% EXO:>95% ] c- and B-subunits b7O Mlspalr Exclslon, % DNA Polymerase Actlvlty, CPM x 109-3 158 12.0 j 11.0: 10.0 j 9.0: 8.01 7.0: 6.0 j 5.0 j 4.0 1 3.0: 2.01 1.0:: 0.0 '- 1M GU‘HCI 10 20 Yield Pok Emm [ ------ ] c- and B-subunits O 30 40 50 60 70 Fractlon Number Figure 268 52% 9 to 33 kDa. The NCS cleavage products derived from the B—subunit were 32, 23, 18 and 11 kDa. The cleavage patterns are completely dissimilar, indicating that the a- and B-subunits of Drosophila Pol Y are proteolytically distinct. Thus, the B— subunit does not appear to have been derived from the a-subunit by proteolysis during enzyme purification. 170 Figure 29. Partial proteolytic digestion of the a- and fi-subunits of Drosophila Y polymerase. DNA polymerase Y fraction VI was subjected to proteolysis by Nechlorosuccinimide as described in "Methods" and was denatured and electrophoresed in a 5-15% linear gradient SDS—polyacrylamide slab gel. Protein detection was by silver staining. Lane 1: 20 ng of DNA polymerase y, lane 2 : 130 ng of SDS-PAGE purified DNA polymerase Y d-subunit, lane 3 :.Nechlorosuccinimide digested Pol Y a-subunit, lane 4: 40 ng of SDS-PAGE purified DNA polymerase Y B-subunit, lane 5 :.Nechlorosuccinimide digested Pol Y B-subunit, lane 6 : sample buffer. 171 l 2 3 4 5 6 ~ “ he: ._' $1 t afl ‘ ;: 1 I ' . J ' I Figure 29 _. 180 kDa .__.-._.43kDa 172 DISCUSSION Antiserum has been developed in rabbits by popliteal lobe injection of the near-homogeneous Drosophila mtDNA polymerase. This serum specifically recognizes the denatured 125 kDa and 35 kDa polypeptides in a crude enzyme fraction (~ 8% pure). It also recognizes the native enzyme as demonstrated by immunoprecipitation of Pol Y and by enzyme inhibition studies. The inhibition analyses show that the antiserum inactivates both the DNA polymerase and 3'-5' exonuclease activities in Pol Y. The extent of inactivation of the the two activities is similar at each serum concentration tested. In contrast, no inhibition of the DNA polymerase or 3'-5' exonuclease associated with E. coli DNA polymerase I or its Klenow fragment was observed, nor were the DNA polymerase or 3'~5' exonuclease associated with Drosophila Y polymerase inhibited by the rabbit pre-immune serum. These results, taken together with the specificity of the antiserum demonstrated by the immunoblot analysis, indicate a physical association of the 3'-5' exonuclease with the two subunit Drosophila Y polymerase. Similar studies were carried out with the nuclear a polymerase. Rabbit polyclonal antiserum developed against Drosophila Pol a was shown to recognize the purified enzyme in an immunoblot and to inhibit DNA polymerase activity in a concentration dependent manner (240). Inhibition of DNA polymerase activity did not exceed 85%, an observation similar ‘to that found with the Y polymerase. This might suggest the 173 exclusion of activity neutralizing antibodies by the binding of non-neutralizing antibodies. Subunit-specific rabbit antiserum to the calf thymus a polymerase was used to aid in defining structure-function relationships between its polypeptides (241). Antiserum directed against the 185 and 160 kDa subunits inhibited DNA polymerase activity, while antibodies directed against the 68, 55, and 48 kDa polypeptides inhibited the DNA primase activity. Immunoblot analyses revealed that the 160 kDa polypeptide is derived by proteolysis from the 185 kDa polypeptide, and that the 48 kDa polypeptide is derived from the 68 kDa polypeptide. Drosophila Y polymerase was shown previously to be a heterodimer composed of a 125 kDa catalytic subunit and 35 kDa subunit of unknown function (159). The subunit association of the 3'-5' exonuclease remains unassigned. To define the role of the B-subunit with regard to the DNA polymerase and/or 3'-5' exonuclease activities, subunit separation experiments were pursued. The function of the B-subunit might be discerned by characterizing an isolated and active a-subunit in terms of processivity and fidelity of DNA polymerization, template-primer utilization, substrate specificity, and kinetics and mispair specificity of exonucleolytic hydrolysis. Dissociation of multi-subunit enzymes has been achieved previously. For example, the Drosophila a polymerase subunits were dissociated and separated by velocity sedimentation in the presence of urea (86,191) and ethylene glycol (138). These studies allowed the assignment of the DNA polymerase and the DNA 174 primase to the 182 kDa and the 50 and 60 kDa subunits, respectively. Further, when ethylene glycol was used as the chaotropic agent, a cryptic 3'-5' exonuclease was discovered and its activity assigned to the 182 kDa DNA polymerase subunit. A major difference between the use of urea and ethylene glycol as the denaturant was the relative recovery of DNA polymerase activity, this being 9% and 40%, respectively. Control experiments with the Drosophila Y polymerase indicated that urea would not be an appropriate denaturant for subunit dissociation and separation. For example, incubation in 2.6 M urea for 8 and 24 hours, the time required for subunit separation by gel filtration and velocity sedimentation, respectively, resulted in losses of 50% and 70% of the DNA polymerase activity. Under the same conditions, the recovery of- 3'-5' exonuclease activity was 18% and 8%. On the other hand, control experiments with ethylene glycol suggested that less than a 20% loss of both DNA polymerase and 3'-5' exonuclease activity would be incurred. These recovery levels are consistent with those obtained for DNA polymerase (86%) and 3'- 5' exonuclease (2 95%) following ethylene glycol-glycerol gradient sedimentation. However, neither the a— nor the B- subunit was detected within the gradient fractions, so that no physical evidence regarding possible subunit separation was obtained. Nevertheless, this high recovery of DNA polymerase and 3'-5' exonuclease activities was similar to that observed upon Sephacryl S-200 gel filtration in the presence of 0.4 M Gu- lKZl -- where subunit separation was nOt achieved, suggesting a 175 lack of subunit separation in the velocity sedimentation analysis. Subunit separation studies were pursued further using Gu-HCl as the chaotropic agent. The rationale for this experimentation was the following: (i) Gu-HCl had previously been shown to partially denature the enzyme hexokinase without a complete loss of activity 4238); (ii) significantly more enzyme could be utilized since the 20-fold dilution required for velocity sedimentation in the presence of ethylene glycol could be avoided; (iii) control experiments indicated the Drosophila Y polymerase was more resistant to inactivation by Gu-HCl than urea. Again however, dissociation and separation of the Drosophila Y polymerase subunits with the retention of catalytic activity was not achieved. Gel filtration of Pol Y in the presence of 1.5 M Gu-HCl appeared to allow for separation of the a- and B- subunits. However, the a-subunit was not detected in those fractions containing DNA polymerase activity as expected, but rather in a much retarded elution position suggesting its denaturation. To improve recovery of the Y polymerase while using Gu-HCl as the denaturant, velocity sedimentation was attempted. Here, some partial subunit separation appears to have occurred, but neither DNA polymerase activity nor 3'-5' exonuclease activity was detected in association with the a— or B-subunits. Thus, under the conditions in which the a- and B-subunits could be separated, little or no enzyme activity could be recovered. 176 Drosophila Y polymerase had previously been subjected to SDS-polyacrylamide gel electrophoresis and partially renatured such that the DNA polymerase activity could be assigned to the 125 kDa subunit (159). Further, the activity of several other proteins was recovered following SDS-PAGE and renaturation via the Hager and Burgess method (194,239), suggesting a similar approach could be pursued with Pol Y. However, neither the isolated a-subunit nor the Y polymerase could be renatured following denaturation by this procedure. {The lack of recovery of the Y polymerase activity following denaturation of the enzyme might be due to the lack of those proteins responsible for folding the enzyme into its native conformation in mitchondria. Such proteins have been termed "molecular chaperones" (29,30). Their primary function is to fold imported proteins and assemble them into functional complexes. Such proteins might be required in a in vitro environment to allow proper refolding of the Y polymerase following denaturation. Alternatively, a more "in vivo"-like hydrophobic environment might be required; the Y polymerase requires both salt and detergent for extraction from mitochondria (159), suggesting a possible membrane association. The failure of the subunit separation studies indicates that the cloning and expression of the a— and B-subunit genes will most likely be required to define the subunit association of the 3'—5' exonuclease, and the functional contribution of the B- subunit to Pol Y activity. In this regard, the baculovirus 177 system has the potential to express active multi-subunit enzymes which are nuclear-encoded but have organellar function(245). Although most studies report the native molecular weight of the mitochondrial DNA polymerase to be between 160 and 180 kDa (158,159,160,161), its subunit structure is debatable. The chick embryo Y polymerase was postulated to be a homotetramer with a subunit size of 47 kDa (158). However, a 130 kDa polypeptide of lower abundance could also be correlated with DNA polymerase activity. An alternate subunit structure was not proposed, nor was the catalytic subunit identified. Drosophila Y polymerase was determined to be a heterodimer composed of a 125 kDa (catalytic subunit) and a 35 kDa subunit with 1:1 stoichiometry (159). The porcine liver and X. laevis Y polymerases do not appear to be homogeneous. SDS-PAGE analysis reveals the presence of four polypeptides in the porcine liver enzyme of 120, 55, 50 and 40 kDa (160). The most highly purified form of X. laevis Y polymerase contains seven polypeptides of 140, 100, 85, 55, 50, 40, and 31 kDa (161). However, ethylene glycol-glycerol gradient sedimentation of this fraction and subsequent SDS-PAGE analysis revealed only three polypeptides of 140, 100 (minor) and 40 kDa (164). Further, DNA cross-linking studies indicated that 100 kDa and 55 kDa polypeptides are likely proteolytic products of the 140 kDa polypeptide (161). Thus, the data suggest two subunits of 140 kDa and 40 kDa in X. laevis Y polymerase, although the 40 kDa subunit might be a inactive proteolytic product of the 140 kDa subunit. 178 To demonstrate that the Drosophila Y polymerase is purified in an intact form and to verify its subunit structure, immunoprecipitates of the enzyme from rapidly prepared embryo and mitochondrial extracts were examined. The data presented here demonstrate the immunoprecipitation of 125 and 35 kDa polypeptides by the rabbit immune serum developed against the Drosophila Y polymerase. The molecular weights of these polypeptides correspond to those of the a- and B-subunits of Pol Y. Rabbit pre-immune serum does not immunoprecipitate polypeptides with corresponding molecular weights. Given the specificity of the anti-Pol Y serum determined by immunoblot and activity neutralization studies, it is likely that the 125 kDa and 35 kDa polypeptides immunoprecipitated from crude embryo and mitochondrial extracts are the intact a- and B-subunits of Pol Y. This suggests that the Drosophila mtDNA polymerase is purified to near-homogeneity in an intact form, and that it is comprised of two structurally diStinct subunits. The question of the structural relatedness of the 125 kDa and 35 kDa subunits has been addressed directly. The data presented here demonstrates that the a- and B-subunits are unrelated: upon cleavage by N-chlorosuccinimide, completely different cleavage patterns are produced. This procedure was also used to examine the structural relationships in the KB cell a polymerase (242). However, in that case, polypeptides of 200, 180, 165 and 140 kDa were shown to have virtually identical NCS cleavage patterns. In the same report, the a polymerase 179 subunits of 77, 55 and 49 kDa were shown to be proteolytically distinct. 180 FUTURE STUDIES The contribution to replication fidelity of the 3'-5' proofreading exonuclease associated with the DrOSOphila mtDNA polymerase remains to be determined. Nucleoside monophosphate (NMP) inhibition and use of a specific nucleotide bias termed "next nucleotide effect" during DNA synthesis are the methods generally used to determine the contribution of a proofreading exonuclease to fidelity (151,168,169). Because nucleoside monophosphates are poor inhibitors of the 3'-5' exonuclease of Drosophila Y polymerase, it seems reasonable to examine other approaches, in addition to those described above, to answer this question. These may include differential chemical inactivation of the 3'-5' exonuclease and DNA polymerase activities or the use of an exonuclease-specific neutralizing polyclonal or monoclonal antibodies. The 3'-5' exonuclease activity of Pol Y was abolished following exposure to 1.0 M Gu-HCl for 10 hours during gel filtration. However, the recovery of DNA polymerase activity was ~50%. A control experiment was performed where Y polymerase was incubated in the presence of 0.4, 0.7 and 1.0M Gu-HCl for 24 hours at 0°C. The results indicate that these concentrations inactivate the 3'-5' exonuclease at a 1.6-, 5.8- and 4.4-fold greater rate, respectively, than the DNA polymerase. Fidelity analysis using the opal reversion assay (168) can be used to determine the accuracy of DNA synthesis, after exposure of Pol Y to Gu-HCl and to evaluate the contribution of the 3'-5' exonuclease to fidelity; the rate of inactivation of the DNA 181 polymerase and 3'-5' exonuclease by Gu-HCl must be examined in the same analysis. A second potential means of determining the contribution of the 3'-5' exonuclease to fidelity is by varying the MgClz and KCl concentrations. In the presence of 30 mM KCl the Y polymerase is highly processive, incorporating several thousand nucleotides per binding event (167). Alternatively, in the presence of 120 mM KCl the Y polymerase is quasi-processive, incorporating ~150 nucleotides per binding event. The 3'-5' exonuclease activity exhibits a lower tolerance to increasing concentrations of MgClz relative to the DNA polymerase activity. For example, in the presence of 10 and 15 mM MgC12, the ratio of 3'-5' exonuclease to DNA polymerase activity is 2- and 3-fold less than at 0.5 mM MgC12. A titration of MgClg was also performed under 30 mM KCl conditions. DNA polymerase activity was 50% of its optimum at 0.25 mM MgC12. However, under the same reaction conditions the 3'-5' exonuclease activity is <4% of its optimal. Thus, it should be possible to determine the contribution of the 3'-5' exonuclease to fidelity under conditions of both high and low processivity by altering the MgC12 concentrations at 30 and 120 mM KCl. Using the specific polyclonal antiserum to Drosophila Y polymerase, cloning of the nuclear DNA sequences encoding the two subunits of Pol Y has been initiated in our laboratory. A Drosophila melanogaster embryo cDNA library was constructed in a A phage expression vector, Agtll, and was provided by the laboratory of Dr. Tao-shih Hsieh of Duke University. This 182 library has been used previously to obtain a full length cDNA clone of the gene encoding topoisomerase II (243). Once identified, the cDNA clones encoding the a— and B-subunits of Pol Y will be expressed in the baculovirus expression system (244,245). This system has been exploited to produce proteins involved in Herpes simplex virus (HSV) replication (246,247). In the case of the Herpes-encoded helicase-primase, a functional enzyme was produced by triple infection of Spodoptera frugiperda cells (preferred host cell for baculovirus) with three recombinant baculoviruses carrying the genes encoding the three subunits of the helicase-primase. The in vivo assembled recombinant enzyme possesses the DNA-dependent ATPase, DNA 'helicase and DNA primase activities that are virtually identical to those contained in the enzyme isolated from HSV-l-infected cells. This suggests it is possible to express a recombinant and functionally competent Drosophila Y polymerase, as well as the individual Pol Y subunits. The recovery of a DNA polymerase and 3'-5' exonuclease competent recombinant Y polymerase and the a-subunit in a DNA polymerase competent form should provide a wealth of information. First, characterization of the recombinant Y polymerase and comparison to native Pol Y'must be performed. This would ensure the recombinant enzyme is enzymatically equivalent to the native enzyme. Second, enzymatic characterization of the isolated a-subunit in terms of template- primer specificity, processivity, fidelity of DNA synthesis, ligand binding affinities, and 3'-5' exonuclease/proofreading 183 activity could be examined. Following comparison to those characteristics of the Pol Y holoenzyme these studies might indicate the functional contribution of the B-subunit. Third, the subunit association of the 3'-5' exonuclease might also be assigned. The expression of the Y polymerase in the baculovirus system will provide a means of obtaining larger quantities of enzyme. The production of recombinant proteins with this system has resulted in levels of 1 to 500 mg/liter of cells (245). The production of larger quantities of Pol Y will aid in the production of monoclonal antibodies and subunit-specific polyclonal antisera. The development of a rapid and efficient immunoaffinity purification procedure for the Drosophila Y polymerase might substantially increase the yield of the enzyme during purification. This could be applied to purification of the native and recombinant Drosophila mtDNA polymerases. Further, a battery of monoclonal antibodies would be useful for epitope mapping, in combination with enzyme neutralization studies. Structure-function analysis of the Drosophila Y polymerase will be enhanced by the acquisition of the DNA sequence of its two subunits. First, amino acid sequence comparisons between the Y polymerase and other eukaryotic and prokaryotic DNA polymerases can be made (248,249). This could provide information suggesting the location of the ligand binding domains and the enzyme active sites. 184 Ligand binding site analysis would be more feasible when milligram quantities of enzyme are available. By performing experiments similar to that for Y polymerase from X. laevis oocytes (160), the DNA binding domain of the Drosophila Y polymerase can be radiolabeled. Subsequent partial and complete proteolysis of the labeled polypeptides could allow for the localization of the DNA binding domain relative to a peptide map. The dNTP binding site could also be localized by analyses similar to those used with E. coli DNA polymerase I (178,179) and terminal deoxynucleotidyltransferase (250). The assembly of the enzyme at a template-primer terminus is also unknown. Using DNA primers with cross-linkable aryl azide residues at specific positions, structural data can be obtained to reveal how the Pol Y a- and B-subunits are positioned on the primer-template. Limited primer extension will allow for site specific cross-linking to be performed over a large array of positions. Similar analysis has been performed with the T4 DNA polymerase (251). The completion of these objectives will provide a greater understanding of the mechanism of DNA synthesis by mitochondrial DNA polymerase. While cloning and overexpression of the Drosophila Y polymerase will greatly facilitate structure- function analysis, some of these experiments can be pursued under the present circumstances. In any case, many experiments will require the use of the purified native enzyme as a control for a those in which recombinant enzyme is used. 185 References 1. Lehninger, A.L. (1964) The Mitochondrion The Benjamin Co. Inc. New York, 263pp 2. Tzagoloff, A. (1981) Afitochondria Plenum Press, New York 342pp 3. Ernster, L. and Schatz, G. (1981) J.Cell Biol. 91, 227—255 4. Hatefi, Y., Haavik, A.G., and Griffiths, D.E. (1962) J. Biol. Chem. 237, 1676-1681 5. Ernster, L. and Kuylenstierna, B. (1970) Membranes of Mitochondria and Chloroplasts (E. Racker ed.), Van Nostrand Reinhold, New York, pp. 172-212 6. Parsons, D.E., Williams, G.R., Thompson,w., Wilson,D., and Chance, B.(1967) Mitochondrial Structure and Compartmentation (E. Quaagliarillo, S. Papa, E.C. Slater and J.M. Trager, eds.) Adriatica Press, Bari, pp. 29-73 7. Koch, G., (1976) J. Biol. Chem. 251, 6097-6107 8. Katan, M.B., Van Harten-Loosbrom, N. and Groot, C.S.P. (1976) Eur. J. Biochem. 70, 409-417 9. Tzagoloff, A. and Meagher, P. (1971) J. Biol.Chem. 246, 7328-7336 10. Dujon, B. (1981) In Molecular Biology of the Yeast Saccharomcyes. Life Cycle and Inheritance, ed. J.N. Strathern, E.W. Jones, J.R. Broach, pp. 505-635. Cold Spring Harbor, NY 11. Lang, B.F., Ahne, F. and Boneen, L, (1985) J. Mel. Biol. 184, 353-366 12. Anderson, S., Bankier, A.T., Barrell, B.G., deBruijn, M.J.L., Coulson, A.R.,Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Sanger,F., Schreier, P.H., Smith, A.J.H., Staden, R. and Young, 1.6. (1981) Nature 290, 457-465 13. Bibb, M.J., Van Etten, R.A., Wright,C.T., Walberg, M.W., and Clayton, D.A. (1981) Cell 26, 167-180 14 Clary, D.O. and Wolstenholme, D.R. (1985) J..Mo1. Evol. 22, 252-271 15. Fox, T.D. and Leaver, C.J. (1981) Cell 26, 315-323 16. Slonimski, P.P. and Tzagoloff, A. (1976) Eur. J. Biochem. 61, 27-41 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 186 Hudspeth, M.E.8., Ainly, W.M., Shamard, D.S., Butow, R.A., Grossman, L.I. (1982) Cell 30, 617-626 Schatz, G., Groot, G.S.P., Mason, T.L., Rouslin, W., Wharton, D.C. and Saltzgaber, J. (1972) Fed. Proc. 31, 21-29 Poutre, C.G. and Fox, T.D. (1987) Genetics 115, 637-647 Costanzo, M.C., Seaver, E.C. and Fox, T.D. (1986) EMBO J. 5, 3637-3641 Costanzo, M.C. and Fox, T.D. (1988) Proc. Natl. Acad Sci. U.S.A. 85, 2677-2681 Schatz, G. and Mason, T.L. (1974) Ann. Rev.Biochem. 43, 51-87 Attardi, G. and Schatz, G.A. (1988) Ann. Rev. Cell Biol. 4, 289-333 Hartl, R.-U. and Neupert, W. (1990) Science 247, 930-938 Pfaner, N. and Neupert, W. (1990) Ann. Rev. Biochem. 59, 331-353 Kiebler, M. (1990) Nature 348, 610-616 Kang, P.-J., Osterman, J., Shilling, S., Neupert, W., Craig, E.A. and Pfanner, N. (1990) Nature 348, 137-143 Roise, D. and Schatz, G. (1988) J. Biol. Chem. 263, 4509- 4511 Hemmingsen, S.M. (1989) Trends Biochem. Sci. 14, 339 Ellis, R.J. (1990) Semin. Cell Biol. 1, 1 Lingist, S. and Craig, E.A. (1988) Ann. Rev. Genet. 22, 632-677 Hemmingsen, S. M. and Ellis, R.J. (1986) J. Plant Physiol. 80, 269-276 Cheng, M.Y., Hartl, F.-U., Martin, J., Pallock, R.A., Kalousek, F., Neupert, W., Hallberg, E.M., Hallberg, R.L. and Horwich, A.L.(1989) Nature 337, 620-625 Osterman, J., Horwich, A.L. Neupert, W., and Hartl, F.-U. (1989) Nature 341, 125-130 Werner-Washburne, M., Stone, D.E., and Craig, E.A. (1987) Mel. Cell. Biol. 7, 2568-2577 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 187 Craig, E.A., Kramer, J, and Kosic-Smithers, J. (1987) P:::. Natl. Acad. Sci. U.S.A. 84, 4156-4160 Nass, S., and Nass, M.M.K. (1963) J. Cell Biol. 19, 613-613 Schatz, G., Haslbrunner, E. and Tuppy, H. (1964) Biochem. Biophys. Res. Commun. 15, 127-132 Clayton, D.A. (1982) Cell 28, 693-705 Bogenhagen, D. and Clayton, D.A. (1977) Cell 11, 719-727 Altman, P.L. and Katz, D.D. (1976) Biological Handbooks, I Cell Biology. pp. 217-219, Federation of American Studies for Experimental Biology, Bethesda, MD Cordonnier, A.M., Dunon-Bluteau, D. and Brun, G. (1987) Nucleic Acids Res. 15, 477-489 VanTuyle, C.B., and Pavco, P.A. (1985) J. Cell Biol. 100, 251-257 Potter, C.A., Fostel, J.M., Berninger, M., Pardue, M.L. and Cech, T. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 4118- 4122 Battey, J. and Clayton D.A. (1980) J. Biol. Chem. 255, 11599-11606 Ojala, D., Merkel, C., Gelfand, R., and Attardi, G.(l980) Cell 22, 393-403 Robberson, D.L., Clayton, D.A. and Morrow, J.F. (1971) Proc. Natl. Acad. Sci. U.S.A 71, 4447-4451 Arnberg, A., van Brugger, G.S.J., ter Schegget, J. and Borst, P. (1971) Biochem. Biophys. Acta. 246, 353-357 Kaguni, L.S. and Clayton, D.A. (1979) Proc. Natl. Acad. Sci. U.S.A. 79, 983-987 Chang, D.D. and Clayton, D.A. (1984) Cell 36, 635-643 Tapper, D.P. and Clayton, D.A. (1981) J. Biol.Chem. 256, 5109-5115 Walberg, M.W. and Clayton, D.A. (1983) J. Biol Chem. 258, 1268-1275 Topper, J.N. and Clayton, D.A. (1989) Mel. and Cell Biol. 9, 1200-1211 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 188 Goddard, J.M. and Wolstenholme, D.R. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 3886-3890 Rubenstein, J.L.R., Brutlag, D. and Clayton, D.A. (1977) Cell 12, 471-482 Klukas, C.K. and Dawid, I.B. (1976) Cell 9, 615-625 Wolstenholme, D.R. and Fauron, C.M.-R. (1976) J. Cell Biol. 73, 279-286 Polan, M.L., Friedman, S., Gall, J.F. and Gehring, W. (1973) J. Cell. Biol. 56, 580-589 Bultmann, H. and Laird, C.D. (1973) Biochim. Biophys. Acta. 299, 196-209 Peacock, W.J., Brutlag, D., Goldring, E., Appeals, R., Ninton, C. and Lindsley, D.C. (1974) Cold Spring Harbor Symp. Quant. Biol. 38, 405-415 Fauron, C.M.-R. and Wolstenholme, D.R. (1976) Proc. Natl. Acad. Sci. U.S.A. 73, 3623-3627 Wolstenholme, D.R., Goddard, J.M. and Fauron, C.M.-R. (1983) Replication of Viral and Cellular Genomes (Becker,B., ed.) pp. 131-148, Martinus Nijhoff Publishing, Boston Goddard, J.M. and Wolstenholme, D.R. (1980) Nucleic Acids Res. 8, 741-757 delete delete Anderson, S., de Bruijn, M.B.L., Coulson, A.R., Eperon, I.C., Sanger, P., and Young, I.G. (1982) J1Mbl.Biol. 156, 683-717 Montoya, J., Ojala, C.L. and Attardi, G. (1981) Nature Chomyn, A., Mariottini, P., Gonzalez-Cadavid, N., Attardi, G., Strong, D., Trovato, D., Riley, M. and Doolittle, R.F. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 5535-5539 Chomyn, A., Mariottini, P., Cleeter, M.W.J., Ragan, I., Matsuno-yagi, A., Hatefi, Yo., Doolittle, R.F. and Attardi, (1985) Nature 314, 592-597 Hare, J.F., Ching, E., and Attardi, G. (1980) Biochemistry 19, 2023-2030 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 189 Barrell, B.G., Bankier, A.T. and Drouin, J. (1979) Nature 282, 189-194 Barrell, B.G., Anderson, S., Bankier, A.T., de Bruijn, M.H.L., Chen, E., Coulson, A.T., Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Walker, J.E., Sanger. Schrier, P.H., Smith, A.J.H., Staden, R., and Young, 1. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 3164-3166 (I) ”l Luck, D.J. L. (1965) J. Cell Biol. 24, 461-470 Nossal, N.G. (1983) Ann. Rev. Biochem 53, 581-615 Challberg, M.D. and Kelly, T.J. (1989) Ann. Rev. Biochem. Kornberg, A. and Baker, T.A. (1991) DNA Replication 2nd edition, W.H. Freeman and Co., San Francisco Funnell, B.E., Baker, T.A. and Kornberg, A. (1987) J. Biol. Chem. 262, 10327-10334 Kung, F.-C. and Glaser, D.A. (1978) J. Bacteriol. 133, 755-762 Sekimizu, K., Bramhill, D. and Kornberg, A. (1987) Cell 50, 259-265 . ‘ Fuller, R.S., Funnell, B.E. and Kornberg, A. (1984) Cell 38, 889-900 _Bramhill, D. and Kornberg, A. (1988) Cell 52, 743-755 Kobori, J.A. and Kornberg, A. (1982) J. Biol. Chem. 257, 13770-13775 Whale, E., Lasken, R.S. and Kornberg, A. (1989) J. Biol. Chem. 264, 2463-2468 Kornberg, A. (1988) J. Biol. Chem. 263, 1-4 McHenry, C.S. Flower, A.M. and Hawker, J.R. (1988) in Cancer Cells/Eukaryotic DNA Replication, eds. Kelly, T. and Stillman, B. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.), Vol.6, pp. 35-41 Kaguni, L.S., Rossignol, J.M., Conaway, R.C., Banks, G.R., and Lehman, I.R. (1983) J. Biol. Chem. 258, 9037-9039 Cha, T.-A. and Alberts, B.M. (1989) J.Biol.Chem. 264, 12220-12225 Hurwitz, J., Dean, F.B., Kwong, A.D., and Lee, S.-H. (1990), J. Biol. Chem. 265, 18043-18046 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 190 Li, J. and Kelly, T. (1984) Proc. Natl. Sci. U.S.A. 58, 6973-6977 Borowiec, J.A., Dean, F.B., Bullock, P.A. and Hurwitz, J. (1990) Cell 60, 181-184 Mastrangelo, I.A., Hough, P.B., Wall, J.S., Dodson, M., Dean, R.B. and Hurwitz, J. (1989) Nature 338, 658-662 Deb, S.P. and Tegtmeyer, P. (1987) J. Virol. 61, 3649-3654 Dean, F.B., Bullock, P., Murakami, Y., Wobbe, C.R., Weissbach, L., and Hurwitz, J.,(l987) Proc. Natl. Acad. Sci. U.S.A. 84, 16-20 Wold, M.S. and Kelly, T.J. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 2523-2527 Virship, D.M. and Kelly T.J. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 3584-3588 Wiekowski, M., Schwartz, M.W. and Stahl, J. (1988) J. Biol. Chem. 263, 436-442 Weinberg, D.H., Collins, K.L., Semancek, P., Russo, A., Wold,M.S., Virshup, D.M. and Kelly, T.J. (1990) Proc. Natl. Acad.Sci. U.S.A. 87, 8692-8696 Ishimi, Y., Claude, A., Bullock, P. and Hurwitz, J.(1988) J. Biol. Chem. 263, 19723-19733 Yang, L., Wold, M.S., Li, J.J.,Kelly, T.J. and Liu, L.F., (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 950-954 Tsurimoto, T. and Stillman, B. (1989) MOl.Cell Biol. 9, 609-619 Eki,T. and Hurwitz, J. (1991) J. Biol Chem. 266, 3087-3100 Tsurimoto, T. and Stillman, B. (1989) EMBO J. 8, 3883-3889 Tsurimoto, T. and Stillman, B. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 1023-1027 delete Hixon, J.E., Wong, T.W., and Clayton, D.A. (1986) J. Biol. Chem. 261, 2384-2390 Chang, D.D., Hauswirth, W.W., and Clayton, D.A. (1985) EMBO J. 4, 1559-1567 Walberg, M.W. and Clayton, D.A. (1983) J. Biol. Chem. 258, 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 191 1268-1275 Chang, D.D. and Clayton, D.A. (1986) Mol. Cell. Biol. 6, 3253-3261 Chang, D.D. and Clayton, D.A. (1987) Science 235, 1178- 1184 Chang, D.D. and Clayton, D A., (1987) EMBO J. 6, 409-417 Chang, D.D. and Clayton, D.A. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 351-355 Klukas, C.K., and Dawid, I.B. (1976) Cell 9, 615-625 Clary, D.O. and Wolstenholme, D.R. (1987) J. M01. Evol. 25, 116-125 Wolstenholme, D.R., Goddard, J.M. and Fauron, C.M.-R (1979) In: Proc. 8th Ann. I.C.N.-U.C.L.A. Symp. on Mblecular and Cellular Biology: Extrachromosomal DNA, Vol.1 (eds. D. Cummings, P. Borst, I.B. Dawid, S. Weisman, and C.F.Fox), Academic Press, New York Bebenek, K., Joyce, C.M., Fitzgerald, M.P. and Kunkel, T.A., (1990) J. Biol. Chem. 265, 13878-13887 Joyce, C.M., Kelley, W.S. and Grindley, N.D.F. (1982) J. Biol. Chem. 257, 1958-1964 McHenry, C.S. and Kornberg, A. (1977) J. Biol. Chem. 252, 6478-6484. McHenry, C.S. (1988) Ann. Rev. Biochem. 57, 519-550 Johanson, K.0., Haynes, T.E., McHenry, C.S. (1986) J. Biol. Chem. 261, 11460-11465 Fersht, A.R. and Knill-Jones, J.W. (1983) J. Mel. Biol. 165, 633-654 Welch, M.M. and McHenry, C.S. (1982) J. Bacteriol. 152, 351-356 Burgers, P.M.J., Kornberg, A. and Sakakibara, Y. (1981) Proc. Natl. Acad. Sci U.S.A. 78, 5391-5395 Fay, P.J., Johanson, K.O., McHenry, C.S. and Bambara, R.A. (1982) J. Biol. Chem. 257, 5692-5699 Griep, M.A. and McHenry, C.S. (1988) Biochemistry 27, 5210-5215 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 192 Scheuermann, R., Tam, S., Burgers, P.M.J., Lu, C., and Echols, H. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 7085- 7089 Flower, A.M. and McHenry, C.S. (1990) Proc. Natl. Acad. Sci U.S.A. 87, 3713-3717 Kodaira, M., Biswas, S.B. and Kornberg, A. (1983) Mol. Genetics 192, 80-86 Maki, S. and Kornberg, A. (1988) J. Biol. Chem. 263, 6547- 6554 Tsuchihashi, Z. and Kornberg, A. (1989) J. Biol. Chem. 264, 17790-17795 Hubscher, U., and Kornberg, A., (1980) J. Biol. Chem. 255, 11698-11703 McHenry, C.S. (1982) J. Biol. Chem. 257, 2657-2663 Burgers, P.J.M. and Kornberg, A. (1982) J. Biol. Chem. 257, 11468-11473 Burgers, P.J.M. and Kornberg, A. (1983) J. Biol. Chem. 259, 7669-7675 Ohashi, M., Taguchi, T. and Ikegami, S. (1978) Biochim. Biophys. Res. Commun. 82, 1084-1090 Wright, G.E., Baril, E.F. and Brown, N.C. (1980) Nucleic Acids Res. 8, 99-105 Edenberg, H.J., Anderson, S. and DePamphilis, M.L. (1978) J. Biol. Chem. 253, 3278-3280 Kaguni, L.S. and Lehman, I.R. (1988) Biochimica Biophysica Acta 950, 87-101 Cotterill, S.M., Reyland, M.B., Loeb, L., and Lehman, I.R. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5635-5639 Fisher, P.A., Chen, J.T. and Korn, D.A. (1981) J. Biol. Chem. 256, 133-141 Fisher, P.A. and Korn, D.A. (1979) J. Biol. Chem. 254, 11040-11046 Fisher, P.A. and Korn, D.A. (1981) Biochemistry 20, 4560- 4569 Reyland, M.E. and Loeb. L.A. (1987) J. Biol. Chem. 262, 10824-10830 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 193 Cotterill, S. Chui, G. and Lehman, I.R. (1987) J. Biol. Chem. 262, 16105-16108 Tseng, B.Y. and Ahlem, C.N. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 2342-2346 Chiu, R.W. and Baril, E.R. (1975) J. Biol. Chem. 250, 7951-7957 Donaldson, R.W. and Gerner, E.W. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 759-763 Tokui, T. Ingak, M., Nishizawa, K., Yatani, R., Kusagawa, M., Ajiro, K., Nishimoto, Y., Date, T. and Matsukage, A. (1991) J. Biol. Chem. 266, 10820-10824 Byrnes, J.J. (1985) Biochem. Biophys. Res. Commun. 132, 628-634 Lee, M.Y.W.T., Toomey, N.L. and Wright, G.E. (1985) Nucleic Acids Res. 13, 8623-8630 Lee, M.Y.W.T. and Toomey, N.L. (1987) Biochemistry 26, 1076-1085 Kunkel, T.A., Sabatino, R.D. and Bambara, R.A. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 4865-4869 Burgers, P.M.J., Bambara, R.A., Campbell, J.A., Chang, L.M.S. Downey, K.M., Hubscher, U., Lee, M.Y.W.T., Linn, S.M., So, A.G., and Spardari, S., (1990) Eur. J; Biochem. 191, 617-618 Wang, T.S.-F. (1991) Ann. Rev. Biochem. 60, 513-522 Hamatake, R.K., Hasegawa, H., Clark, A.B., Bebenek, K. and Kunkel, T.A. (1990) J. Biol. Chem. 265, 4072-4083 Abbots, J., Dibyendu, N., SenGupta, D., Zmudzka, B.A., Widen, G.G., Notario, V. and Wilson, S.H. (1988) Biochemistry 27, 901-909 . Kunkel, T.A.(1985) J. Biol. Chem. 260, 5787-5796 McBride, O.W., Zmudzka, B.A. and Wilson, S.H. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 503-507 Masamitsu, Y., Matsukage, A., and Takahashi, T. (1980) J. Biol. Chem. 255, 7002-7009 Wernette, C.M. and Kaguni, L.S. (1986) J. Biol. Chem. 261, 14764-14770 Insdorf, N.F. and Bogenhagen, D.F. (1989) J. Biol. Chem. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 194 264, 21491-21497 Kunkel, T.A. and Mosbaugh, D.W. (1989) J. Biol. Chem. 28, 988-995 Kunkel, T.A. and Soni, A. (1988) J. Biol. Chem. 263, 4458- 4459 Kaguni, L.S. and Olson, M.W. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 6469-6473 Insdorf, N.F. and Bogenhagen, D.F. (1989) J. Biol. Chem. 264, 21498-21503 Brown,. W.M., George, M.,Jr. and Wilson, A.C. (1979) Proc. Natl. Acad. Sci. U.S.A. 76,1967-1971 Clayton, D.A., Doda, J.N. and Friedberg, E.C. (1974) Proc. Natl. Acad. Sci. U.S.A. 71, 2777-2781 Wernette, C.M. Conway, M.C. and Kaguni, L.S. (1988) Biochemistry 27, 6046-6054 Kunkel, T.A. (1985) J. Biol. Chem. 260, 5787-5796 Kunkel, T.A. (1988) Cell 53, 837-840 Kunkel, T.A., Schaaper, R.M., Beckman, R.A. and Loeb, L.A. (1981) J .Biol. Chem. 256, 9983-9987 Maki, H. and Kornberg, A. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 4389-4392 Kunkel, T.A., Loeb, L.A. and Goodman, M.F. (1984) J. Biol. Chem. 259, 1539-1545 Sinha, N.K. (1987) Proc. Natl. Acad. Sci U.S.A. 84, 915- 919 Skarnes, W., Bonin, P. and Baril, E (1986) J. Biol. Chem. 261, 6629-6636 Joyce, C.M. (1989) J. Biol. Chem. 264, 10858-10866 Das, S.K. and Fujimura, R.K. (1980) J. Biol. Chem. 255, 7149-7154 Ollis, D.L., Brick, P., Hamlin, R., Xuong, N.G., and Steitz, T.A. (1985) Nature 313, 762-766 Joyce, C.M., Ollis, D.L., Rush, J., Steitz, T.A., Konigsberg, W.H.,and Grindley, N.D.F. (1985) in Protein Structure, Folding and Design (UCLA Symposia on Molecular 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 195 and Cellular Biology, Vol 32,) (D. Oxender, ed.), Alan R. Liss pp. 197-205 Basu, A. and Modak, M.J. (1987) Biochemistry 26, 1704- 1709 Polesky, A.H., Steitz, T.A., Grindley, N.D.F. and Joyce, C.M. (1990) J. Biol. Chem. 265, 14579-14591 Derbyshire, V., Freemont, P.S., Sanderson, M.R., Beese,L., Friedman, J.M., Joyce, C.M. and Steitz, T.A. (1988) Science 240, 199-201 Kuwabara, N.and Uchida, H. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 5764-5767 LaDuca, J.J., Crute, J.J., McHenry, C.S., Bambara, R.A. (1986) J. Biol. Chem. 261, 7550-7557 Johanson, K.O. and McHenry, C.S. (1980) J. Biol. Chem. 255, 10984-10990 Engstrom, J., Wong, A. and Maurer, R. (1986) Genetics 113, 499- 515 O'Donnell, M.E. (1987) J. Biol. Chem. 262, 16558-16565 Wong, S.W., Wahl, A.F., Yuan, P.-M., Arai, N., Pearson, B.E., Arai, K.-I., Korn, D., Hunkapiller, M.W. and Wang, T.S.-F. (1988) EMBO J. 7,37-48 Zmudzka, B.A., SenGupta, D., Matsukage, A., Cobianchi,F., and Wilson, S.H. (1986) Proc .Natl. Acad. Sci. U.S.A. 83, 5106-5110 Fansler, 8.3. and Loeb, L.A. (1974) Methods. Enzymol. 29, 53-70 McHenry, C.S. and Crow, W. (1979) J. Biol. Chem. 254, 1748-1753 Kaguni, L.S., Rossignol, J.-M., Conaway, R.C. and Lehman, I.R. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 2221- 2225 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254 Laemmli, U.K. (1970) Nature 227, 680-685 Hager, D.A. and Burgess, R.R. (1980) Anal. Biochem. 109, 76-86 Wray, W., Boulikas, T., Wray, V.P. and Hancock, R. (1981) Anal. Biol. Chem. 118, 197-203 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 196 Lishewe, M.A. and Ochs, D. (1982) Anal. Biochem. 127, 453- 457 Loeb, L.A. and Kunkel, T.A. (1982) Ann. Rev. Biochem. 51, 429-457 Dawid, I.B. (1972) Devel. Biol. 29, 139-151 Kaguni, L.S., Wernette, C.M., Conway, M.C. and Yang- Cashman, P. (1988) in Cancer Cells: Eukaryotic DNA Replication, eds. Kelly, T. and Stillman, B. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) V61. 6, pp.425-432 Kunkel, T.A. and Loeb, L.A. (1981) Science 213, 765-766 Krauss, S.W. and Linn, S. (1980) Biochemistry 19, 220-228 Krauss, S.W. and Linn, S. (1982) Biochemistry 21, 1002- 1009 Kunkel, T.A. and Alexander, P.S. (1986) J. Biol. Chem. 261, 160-166 Byrnes J.J., Downey, K.M., Black, V.L. and So, A.G. (1976) Biochemistry 15, 2817-2823 Fersht, A.R., Knill-Jones, J.W. and Tsui, W.-C. (1982) J. Mo1. Biol. 156, 37-51 Echols, H., Lu, C. and Burgers, P.M.J. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 2189-2192 Ottiger, H.-P. and Hubscher, U. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 3993-3997 Vishwanatha, J.K. Coughlin, S.A., Wesolowski-Owen, M. and Baril, E.F. (1986) J. Biol. Chem. 261, 6619-6628 de Bruijn, M.H.L. (1983) Nature 304, 234-241 Cotterill, S.M., Reyland, M.E., Loeb, L.A. and Lehman, I.R. (1988) in Cancer Cells: Eukaryotic DNA Replication, eds. Kelly, T. and Stillman, B. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) Vol. 6, 367-371 Kaguni, L.S., DiFrancesco, R.A. and Lehman, I.R. (1984) J. Biol. Chem. 259, 9314-9319 Wolstenholme, D.R. and Clary, D.O. (1985) Genetics 109, 725-744 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 197 Reyland, M.E., Lehman, I.R. and Loeb, L.A. (1988) J. Biol. Chem. 263, 6518-6524 Drake, J.W. (1969) Nature, 221, 1132-1134 Mildvan, A.S. (1974) Ann. Rev. Biochem. 43, 357-399 Springgate, C.F. and Loeb, L.A. (1973) Proc. Natl. Acad. Sci. U.S.A. 70, 245-249 Hall, Z.W. and Lehman, I.R. (1968) J. Mol. Biol. 36, 321- 332 Que, B.G., Downey, C.M. and So, A.G. (1978) Biochemistry 17, 1603-1607 Fry, M. and Loeb, L.A. (1986) Animal Cell DNA Polymerases, CRC Press Inc., Bocco Raton, Fl. Brutlag, D. and Kornberg, A. (1972) J. Biol. Chem. 247, 241-248 Scheuerman, R.J. and Echols, H. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 7747-7751 Das, S.K. and Fujimura, R.K. (1980) Nucleic Acids Res. 8, 657-671 Abbotts, J. and Loeb, L.A. (1985) Nucleic Acids Res. 13, 261-274 Hubermann, J.A. and Kornberg, A. (1970) J. Biol. Chem. 245, 5326-5334 Byrnes, J.J., Downey, K.M., Que, B.G., Lee, M.Y.W., Black, V.C. and So, A.G. (1977) Biochemistry 16, 3740-3746 delete delete Matsukage, A., Bohn, E.W. and Wilson, S.H. (1975) Biochemistry 14, 1006-1020 Fisher, P.A., Wang, T.S.-F. and Korn, D. (1979) J. Biol. Chem. 254, 6128-6137 Hammond, R.A., Brynes, J.J. and Miller, M.R. (1987) Biochemistry 27, 6817-6824 Sabatino, R.D. and Bambara, R.A. (1988) Biochemistry 27, 2266-2271 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 198 Setlow, P., Brutlag, D. and Kornberg, A. (1972) J. Biol. Chem. 247, 224-231 Griep, M.A. and McHenry, C.S. (1989) J. Biol. Chem. 264, 11294-11301 Syvaoja, J. and Linn, S. (1989) J. Biol. Chem. 264, 2489- 2497 Reems, J.A., Griep, M.A., McHenry, C.S. (1991) J. Biol. Chem. 266, 4878-4882 Mendelmann, L.V., Boosalis, M S., Petruska, J. and Goodmann, M.F. (1989) J. Biol. Chem. 264, 14415-14423 Mizrahi, V., Benkovic, P. and Benkovic, S.J. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 5769-5773 White, T.K. and Wilson, J.E. (1989) Arch. Biochem. Biophys. 274, 375-393 Burton, Z.F., Killeen, M., Sopta, M., Ortolan, L.G. and Greenblatt, J. (1988) Mel. Cell. Biol. 8, 1602-1613 Sauer, B. and Lehman, I.R. (1982) J. Biol. Chem. 257, 12394-12398 Holmes, A.M., Cheriathundam, E., Bollum, F.J. and Chang, L.M. (1987) J. Biol. Chem. 261, 11924-11930 Wong, S.W., Paborsky, L.R., Fisher, P.A., Wang, T.S.-F. and Korn, D.A. (1986) J. Biol. Chem. 261, 7958-7968 Nolan, J.M., Lee, M.P., Wyckoff, E., and Hsieh, T. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 3664-3668 ‘ Smith, G.A., Summers, M.D., and Fraser, M.J. (1983) Mel. Cell. Biol. 3, 2156-2165 Lucknow, V.A. and Summers, M.D. (1988) Bio/Technology 6, 47-55 Olivo, P.D., Nelson, N.J. and Challberg, M.D. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 5414-5418 Dodson, M.S., Crute, J.J., Bruckner, R.C. and Lehman, I.R. (1989) J. Biol. Chem. 264, 20835-20838 Bernad, A., Blanco, L., Lazaro, J.M., Martin, G. and Salas, M. (1989) Cell 59, 221-228 Bernad, A., Lazaro, J.M., Salas, M. and Blanco, L. (1990) Proc. Natl. Acad Sci. U.S.A. 87, 4610-4614 199 250. Pandey, V. and Modak, M.J. (1988) J. Biol. Chem. 263, 3744-3751 251. Capson, T.L., Benkovic, S.J ., and Nossal, N.G. (1991) Ce~ 65, 249-258 1}‘l||‘.'l'llv ll ‘11 1 ‘- "71111111111141?