1153 W710 T 3". LIBRARY 1: trims ”China" Stat. University This is to certify that the dissertation entitled Mitochondrial DNA Polymerase From Drosophila Melanogaster Embryos: Purification, Subunit Structure and Template-Primer Utilization Studies presented by Catherine Marie Wernette has been accepted towards fulfillment of the requirements for Ph.D. degree in Biochemistry W5 5. W'- Major professor Date 9/20/88 MSU is an Affirmative Action/Equal Opportunity Institution 0—12771 _____._.____. .h MSU RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from 4—13—- your record. FINES will be charged if book is returned after the date stamped below. MITOCHONDRIAL DNA POLYMERASE FROM DRQS QPHILA MELANSEASTER EMBRYOS: PURIFICATION, SUBUNIT STRUCTURE AND TEMPLATE-PRIMER UTILIZATION STUDIES By Catherine Marie Wemette A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1988 C’s: <41.) 1, r, 7“». 'L"\ ABSTRACT Mitochondrial DNA Polymerase from 12295221113 W Embryos: Purification, Subunit Structure and Template-primer Utilization Studies By Catherine Marie Wemette Mitochondrial DNA polymerase was purified to near homogeneity from early embryos of Mia W (Oregon R). Sodium dodecyl sulfate gel electrophoresis of the highly pru'ified enzyme reveals two polypeptides of 125,000 and 35,000 daltons in a ratio of 1:1. The enzyme has a sedimentation coefficient of 7.6 S and a Stokes radius of 51 A. Taken together, the data suggest that the W DNA polymerase ‘y is a heterodimer. DNA polymerase activity gel analysis has allowed the assignment of the DNA polymerization function to the large enzyme subunit. The mitochondrial DNA polymerase demonstrates a high degree of accuracy in nucleotide incorporation which is nearly identical to that of the replicative DNA polymerase or from Dmsgphila embryos. DNA polymerase y exhibits a remarkable ability to utilize efficiently a variety of template-primers including gapped, multi- or singly-primed natural DNAs and synthetic deoxyribo- or ribohomopolymers. Kinetic experiments and direct physical analysis of DNA synthetic products indicate that the 1212959211113 DNA polymerase ‘y polymerizes nucleotides by a quasi-processive mechanism. The processivity is not affected by the reaction pH (in the range of pH 6-pH 9) or the concentration of divalent cation examined, although these factors do affecr the reaction rate. Inclusion of a heterologous single-stranded DNA binding protein increases 2-3 fold both the rate of DNA synthesis and the processivity of the enzyme on natural DNA but does not result in a completely processive enzyme. In contrast, Dmmhfla DNA polymerase y is processive for several thousand nucleotides under low ionic strength conditions which are suboptimal for DNA synthetic rate. Thus, the unusual template-primer usage behavior under reaction conditions that are optimal for DNA synthetic rate appears to be an intrinsic function of the enzyme in its current highly purified state. However, the catalytic properties of the near- homogeneous Drgsgphila DNA polymerase y are consistent with the i_n vivg requirements for mitochondrial DNA synthesis as described in a variety of animal systems. For my husband and my son. ACKNOWLEDGEMENT I wish to express my gratitude to my major professor, Dr. Laurie. S. Kaguni, for the intellectual effort and financial support she was able to provide during the course of this work. I would also like to extend my appreciation to those who have served on my guidance committee. Dr. John Wilson, Dr. Shelagh Ferguson-Miller, Dr. Ron Davis, Dr. Tom Friedman, Dr. William Deal and Dr. Bob Hausinger have all expressed considerable interest in my project. All of the past and present members of both the Dr. L. S. Kaguni and Dr. J. M. Kaguni laboratories have provided an interesting working environment and a constant source of amusement. Thanks to Dr. Jon. M. Kaguni, Theodore R. Hupp, Phyllis Yang- Cashman, Dr. Deog Su Hwang, Dave Siemieniak, Matthew Olsen, Qingping Wang, Richard Newcomb, Kevin Carr and David Lewis for their friendship and support. I am grateful for the advice I received from Al Smith and Dave Schwab. Dr. Zach Burton generously allowed me use of his Macintosh II. Also, thanks to Tracy K. White, Linda Gregory and Annette Thelen for their friendship and encouragement. Finally, I appreciate the many helpful discussions and encouragement provided by the Society of Disembodied Biochemists. TABLE OF CONTENTS Page List of Tables .................................................................................................. x List of Figures ................................................................................................ xi List of Abbreviations ....................................................................................... xiii Chapterl LITERATURE REVIEW .......................................................... 1 Mitochondrial DNA ................................................................ 2 Replication of Mitochondrial DNA ............................................... 6 Mitochondrial DNA Replication Proteins ....................................... 12 Eukaryotic DNA Polymerases ................................................... 14 Replicative Enzymes-DNA polymerase or and DNA polymerase 6 .............................................. 15 DNA Repair Enzymes-DNA polymerase [3 .......................... 20 Mitochondrial Enzymes--DNA polymerase y ........................ 21 Fidelity of DNA Synthesis .............................................. 24 Processivity of DNA Synthesis ......................................... 30 References .......................................................................... 33 Chapter II. A MITOCHONDRIAL DNA POLYMERASE FROM EMBRYOS OF WW3; PURIFICATION. SUBUNIT STRUCTURE AND PARTIAL CHARACTERIZATION ................... 42 Abstract ............................................................................ 43 Introduction ........................................................................ 44 vi Chapter HI. Experimental Procedures ......................................................... 45 Materials ................................................................... 45 Methods ................................................................... 46 Processing of 2mm embryos ............................ 46 Preparation of Partially Purified Mitochondria ............... 46 DNA polymerase ‘y Assay ...................................... 47 Phosphocellulose Chromatography and Ammonium SulfateFractionation ............................................ 48 DNA-Cellulose Chromatography ............................. 48 Octyl-Sepharose Chromatography ............................ 48 Cibacron Blue-Agarose Chromatography .................... 49 Glycerol Gradient Sedimentation .............................. 49 Gel Filtration ..................................................... 49 Gel Electrophoresis and Protein Transfers ................... 52 Analysis of DNA polymerase Activity in Sign ............... 52 Protein Determinations .......................................... 53 Results ............................................................................ 53 Purification of W DNA polymerase 7 ........................ 53 ' Physical Properties ....................................................... 56 DNA Polymerization in Sim ............................................ 63 Reaction Requirements .................................................. 63 Thermal Stability and Role of DNA, dNTPs and ATP .............. 70 Template-primer Specificity ............................................. 75 Discussion ......................................................................... 77 References .......................................................................... 80 KINETICS, PROCESSIVITY AND FIDELITY OF DNA POLYMERIZATION ............................................................. 82 vii Chapter IV. Abstract ............................................................................ 83 Introduction ....................................................................... 84 Experimental procedures ......................................................... 85 Materials ................................................................... 8 5 Methods ................................................................... 8 6 DNA polymerase y Assay ...................................... 86 Steady State Kinetic Analysis of Template-primer Utilization ........................................................ 87 Quantitative Kinetic Analysis of Processivity ............... 87 Analysis of the Products of DNA Synthesis by Denan Polyacrylamide Gel Electrophoresis .......................... 88 Fidelity of DNA Synthesis ..................................... 88 Results .............................................................................. 89 Template-primer Specificity and Rate of Nucleotide Polymerization ............................................................ 89 Processivity of Nucleotide Polymerization .......................... 105 Fidelity of in m DNA synthesis ................................... 110 Discussion ........................................................................ 112 References ........................................................................ 117 PROCESSIVITY OF THE MITOCHONDRIAL DNA POLYMERASE : TEMPLATE-PRIMER, REACTION CONDITION AND DNA BINDING PROTEIN EFFECTS ................................ 120 Abstract ........................................................................... 121 Introduction ....................................................................... 122 Experimental procedures ........................................................ 123 Materials ................................................................. 123 Methods .................................................................. 124 DNA polymerase y Assay ..................................... 124 Analysis of the Products of DNA Synthesis by Denaturing Gel Electrophoresis ............................................ 124 viii Chapter V. Results ............................................................................ 125 DNA Synthesis on Synthetic Homopolymers Poly (dA)- -oligo (dT)5 and Poly (rA)- 011 go (dT) ............................................... Processivity on a Synthetic Template-primer and Effect of Reaction pH .................................................. 128 Processivity on Natural DNA and Effect of KCl Concentration ............................................................ 131 Processivity on Natural DNA and Influence of Single-stranded DNA Binding Protein .................................................. 137 Effect of Template-primer and Enzyme Concentration on Processivity .............................................................. 142 Comparison of the Processivity of Crude and Highly Purified DNA polymerase 7 ...................................................... 147 Discussion ........................................................................ 150 References ........... ' ............................................................. 1 56 Summary and Perspectives ..................................................... 158 Appendix A ....................................................................... 164 Appendix B ....................................................................... 166 ix Chapter II Chapter III LIST OF TABLES Page DNA polymerase 7 activity during development of D_. melanogaster embryos ........................................................ 54 Purification of DNA polymerase y from 12. W embryos ............................................................................ 55 Template-primer sepcificity of DNA polymerase ‘y from D. W embryos ........................................................ 76 Kinetic parameters of template-primer utilization by DNA polymerase y ................................................................ 93 Kinetic assessment of the processivity of nucleotide polymerization by DNA polymerase 7 .............. 106 Reversion frequency of ¢X 174am3 DNA synthesized in 21119 by DNA polymerase y ........................................................... 1 l l Chapter I l . 2 3 . 4 Chapter II 1 . 2. LIST OF FIGURES Page Structure of the genetic map of W mitochondrial DNA .............. 4 Model for bidirectional synthesis of DNA ....................................... 8 Model for replication of mouse mitochondrial DNA .......................... 10 o X174am3 reversion assay ...................................................... 27 Purification of the Qmsgphfla W DNA polymerase ‘y ............ 51 SDS-polyacrylamide gel electrophoresis of Q. WDNA polymerase 7 ....................................................................... 58 Glycerol gradient sedimentation of mm DNA polymerase y ......... 60 Gel filtration of MI; DNA polymerase y ............................... 62 DNA polymerization by crude and near-homogeneous M3 DNA polymerase y in sign ........................................................ 65 Dependence of DNA polymerase 7 activity on monovalent cation concentration ....................................................................... 67 Determination of Km for dTTP of the crude and near-homogeneous W DNA polymerase y .................................................. 69 Inhibition of the MM W DNA polymerase yactivity by dideoxythymidine triphosphate and N -ethylmaleimide ....................... 72 Thermal stability of DNA polymerase 7 ........................................ 74 xi Chapter III 1 . 2. 6. Chapter IV 1 . 2. Time course of DNA synthesis in template-primer excess ................... 96 Analysis of the products of DNA synthesis on singly-primed tax 174 DNA ....................................................................... 95 Time course of DNA synthesis in enzyme excess ............................. 98 Analysis of the products of DNA synthesis under conditions of template-primer limitation .................................................... 101 Analysis of the products of DNA synthesis in the presence of limiting amounts of singly-primed ¢Xl74 DNA ....................................... 104 Gel analysis of the processivity of DNA polymerase ‘y ...................... 109 Synthesis on synthetic homopolymers and effect of divalent cation ...... 127 Effect of reaction pH on the rate of synthesis and processivity on poly (dA)-oligo (dT) ......................................................... 130 Analysis of the DNA products synthesized by DNA polymerase 7 on poly (dA)-oligo (dT) and the effect of reaction pH ....................... 133 Effect of KCl on the DNA products synthesized by DNA polymerase y ............................................................... 136 Analysis of the DNA products synthesized by DNA polymerase 7 on singly-primed M13 DNA and the effect of single-stranded DNA binding protein ............................................................ 139 Effect of a heterologous single-stranded DNA binding protein on the rate of DNA synthesis and processivity of DNA polymerase 7 on singly-primed M13 DNA ................................... 141 Effect of template-primer concentration on the DNA products synthesized by DNA polymerase y ............................................ 144 Efi‘ect of enzyme concentration on the DNA products synthesized by DNA polymerase y ............................................ 146 Comparison of the DNA products synthesized by other mitochondrial DNA polymerase 7 fractions ................................... 149 xii BSA dzTI'P D-loop D'IT kb mtDNA NEM nt OH, H-strand OL, L-strand PCN A PMSF Pol I Pol a P01 7 RF SSB SDS ABBREVIATIONS bovine serum albumin dideoxy 'ITP displacement loop dithiothreitol hour(s) kilobase(s) mitochondrial DNA N-ethylmaleimide nucleotide(s) mitochondrial origin of DNA synthesis, heavy strand mitochondrial origin of DNA synthesis, light strand proliferating cell nuclear antigen phenylmethylsulfonyl fluoride E. £911 DNA polymerase I DNA polymerase on DNA polymerase y replicative form E. c5211 single-stranded DNA binding protein sodium dodecyl sulfate xiii Chapter I LITERATURE REVIEW i h n ' DN Mitochondrial DNA (mtDNA) comprises less than 1% of the total cellular DNA (1). The density of mitochondrial organelles is uniform throughout the cell cycle and mtDNA can replicate during any phase of the cell cycle in mouse L cells (2). It is not known whether duplication of the organelle and replication of the mitochondrial genome are coordinated. Further, the actual site of mtDNA synthesis within the organelle has not been identified although mtDNA has been isolated in association with protein (3-5) and, in Drosophila, certain sites within a segment of the mitochondrial genome rich in adenine and thymine nucleotides have been shown to be resistant to crossan agents (6). These studies suggest that the mtDNA is associated with protein at discrete sites and is probably localized within and associated with the inner mitochondrial membrane. The majority of metazoan mtDNAs range in size from 14.5-16.5 kilobases (kb). However, the mtDNA of the genus 1213159213113 (Figure l) ranges from 15.7 to 19.5 kb (7). The greater size of the mtDNA of this genus is due to the presence of a single region rich in adenine and thymine nucleotides termed the A+T rich region. This region which comprises 95% adenine and thymine residues was discovered by heat denaturation studies of We mtDNA replication intermediates (8-10) and varies in size between the mtDNA molecules of the melanogaster group from 1-5.1 kb (7); it is approximately 1 kb in all other 912mm species. Further evidence for the existence of the A+T rich region in Drosgphila mtDNAs was obtained from alkaline denaturation (11) and buoyant density studies of mtDNA (10,12,13). Although this A+T rich region varies in size it is located in an identical position in all Dmmphfla species examined relative to the rest of the mtDNA genome (14). A+T rich regions were also identified in yeast mtDNA (15) but have not been found in any other organism. The function of the A+T rich region is unknown although the origin of mtDNA replication for six species of W has been localized to this area by denaturation Figure 1. Structure of the genetic map of Drosoghilg mitochondrial DNA. A+T, A+T rich region of Drgsgphila mitochondrial DNA; 0, origin of mitochondrial DNA replication; R, direction of leading strand DNA replication; sm rRNA, coding region for the small ribosomal RNA subunit; 1g rRNA, coding region for the large ribosomal RNA subunit; NDl, ND2, ND3, ND4, ND4L, ND5, ND6, coding regions for six subunits of the NADH dehydrogenase complex; cyt b, coding region for cytochrome b; COI, C011, C011], coding regions for cytochrome oxidase subunits I, II and III; ATPase 6, coding region for ATPase subunit 6; ATPase 8, coding region for ATPase subunit 8. Small capital letters represent the standard nomenclature for the genes which encode tRN As. The arrow associated with each coding region indicates the direction of transcription. DROSOPHILA MITOCHON DRIAL DNA 5 mapping and analysis of restriction digests of partially replicated mtDNA molecules by electron microscopy (16). The A+T rich region probably does not encode protein due to its low G+C content (~5%), lack of conserved sequences between the various species, and high frequency of termination codons (17). Further, no RNA transcripts which hybridize to this A+T rich region of the mtDNA have been detected in 12325293113 (18). However, promoters for heavy and light strand transcription of mouse mtDNA are located within the replication origin region (19) and it is likely that this area of the genome plays an important role in the control of mitochondrial gene expression. Like the 2mm mtDNA replication origin, the mammalian mtDNA replication origins are variable in length, exhibit little conservation of nucleotide sequence between species and lack any open reading frames (20). The G+C content for human, mouse and Xenopus heir} mtDNAs are 44.3%, 36.7% and 42.9%, respectively (21-23). Outside of the A+T rich region the G+C content for W mtDNA is 25% (17) a considerably lower value. Even though the mammalian mtDNAs have no A+T rich region comparable in size to that of Drosgphila and the G+C content is greater than 30%, the nucleotide sequence of the human HeLa cell mtDNA origin contains 13 consecutive adenine-thymine base pairs beginning 18 base pairs upstream from the origin (24). Also, a run of 14 adenine—thymine base pairs is present within the rat mtDN A replication origin (25). It is possible that these A+T regions may be important for origin function. The human, bovine and mouse mitochondrial genomes have been entirely sequenced and the mitochondrial gene products identified (21,22,26—28). These studies have shown that mtDNA encodes two rRNAs (128 and 16S) and twenty-two tRN As which are necessary and sufficient for mitochondrial protein synthesis. In addition, thirteen mitochondrial gene products have been identified as components of enzyme complexes located in the mitochondrial inner membrane that function in electron transport and oxidative phosphorylation. These include cytochrome b (cyt b), cytochrome oxidase 6 subunits I, II and III (COI-COIII), ATPase subunits 6 and 8, and six components of the respiratory-chain NADH dehydrogenase complex (NDl-ND6). Replication of Mitgghgnm'al DNA Much of the information on DNA replication has been obtained from the extensive biochemical and genetic studies of Escherichia cell and its viruses (29,30). DNA replication is a semi—conservative process that usually initiates at a specific DNA sequence (29) known as the origin of replication (Figure 2). The process is regulated at the point of initiation and requires synthesis of a short primer which is extended by a DNA polymerase. Synthesis may be uni- or bidirectional and proceeds by addition of nucleotide monomers in a 5'-3' direction in a semidiscontinuous manner. This means that synthesis is continuous on one strand (the leading strand) and discontinuous on the other (the lagging strand). While less is known about the replication of the chromosomal or mitochondrial DNA of higher organisms the available data suggest that the biochemical and genetic components may be similar. The information relating to the analysis of the mechanism of mammalian and Mia mitochondrial DNA replication has been derived from electron microscopic studies of mtDNA replication intermediates. Early studies showed that mouse mtDNA was separated into two major bands when isolated on ethidium bromide-cesium chloride buoyant density gradients. The lower band contained predominantly closed circular mtDNAs with a displaced loop structure, while the upper band contained nicked and linear DNA as well as more complex structures, such as expanded and gapped circular DNAs (31). These studies suggested an asymmetric, uni-directional model for semidiscontinuous replication of animal mtDNA with initiation at a unique site (32-35). The scheme proposed as a model for the replication of mammalian mtDNA was determined by electron microscopic examination of the replicative intermediates from mouse L cells (Figure 3). Since the two strands of mammalian mtDNA exhibit a nucleotide Figure 2. Model for the bidirectional replication of DNA. Replication is semidiscontinuous: continuous on the leading strand and discontinuous on the lagging strand. (From: Kornberg, A. (1980) DNA Replication, W. H. Freeman and Co., P. 349.) Figure 3. Model for replication of mouse mitochondrial DNA. Parental H and L- strands are represented by thick and thin solid lines; daughter H and L-strands are represented by thick and thin broken lines. The carat represents a single-strand break in the at daughter molecule. OH, the sequence of the H-strand at the H- strand origin region; OL, the sequence of the L-strand at the L-strand origin region; D- mtDNA, mtDNA with a partially displaced parental H-strand; or, a nicked daughter molecule; [3 Gpc, a gapped daughter molecule. (From: Clayton, D. A. (1983) Cell 28: p. 694.) 10 A 0L / (Hum 11 bias they are easily separated into heavy (H) and light (L) strands by buoyant density gradient centrifugation (36). The predominant form of mtDNA is a covalently closed circle with a displacement loop (D-loop) at the origin of replication of the H—strand. The D-loop is a triple- stranded structure formed by repeated synthesis and degradation of a family of short, single-stranded DNAs complementary to the L-strand, with displacement of the parental H-strand as a single-stranded loop. It is not known whether replication initiates from the short DNA complementary to the L-strand in the D-loop or whether another primer is synthesized and extended; replication proceeds unidirectionally by daughter H- strand synthesis with concurrent displacement of the parental H-strand. The origin of L- strand synthesis is exposed at a point where daughter H—strand synthesis is 67% complete. The exposed nucleotide sequence at the origin of light strand synthesis fomrs a specific secondary structure which appears to be the site of primer synthesis and subsequent chain extension in both the mouse and human systems (37). This highly asymmetric mode of unidirectional synthesis results in the production of two types of daughter molecules: a nicked double-stranded circle and a gapped molecule. When DNA synthesis is completed these molecules are converted to covalently closed circles which are subsequently supercoiled (1). The Mia mtDNA does not have a base bias in its DNA strands and therefore, they cannot be designated as heavy or light (7). Further, it has not been possible to isolate Drosgphila mtDNA containing the D-loop structure (7,11,38). It is not known whether this represents an important difference in the mechanism of mt DNA replication in mm or whether the D400p structures are lost during isolation of the mtDNA due to branch migration (39). It is also clear that in Dmsgphjla leading DNA strand synthesis is generally 87 -99% complete before lagging DNA strand synthesis ensues (16) WWW While examination of mtDNA replication intermediates has allowed the development of models for mtDN A replication there is little information regarding the number and kind of protein components that are required to carry out the replication process in m. The mitochondrial DNA polymerase, which is reviewed later, has been partially characterized because of its central importance to the replication process. Other efforts to characterize enzymatic components have centered around identification of proteins that may function to prime mtDNA synthesis and the mtDNA binding proteins. The 3'-ends of RNAs initiated at the mouse H-strand promoter (74-163 nt upstream) lie within a region near the 5'-ends of nascent displacement loop DNA strands. In fact, the RNA transcript may be covalently linked to the nascent DNA (40). These results indicate that priming of DNA replication at the mouse H-strand origin may be accomplished by a mitochondrial RNA polymerase. However, this has not been proven. A different priming mechanism may be operative at the mouse L-strand origin. Crude mitochondrial protein fractions that can initiate replication on single-stranded DNA templates in m have been isolated from human KB cells (37) and rat liver (41). Both of these enzyme systems are dependent upon rNTPs which are utilized in the synthesis of RNA primers that are 9- 12 nucleotides long. These primers can be extended by crude preparations of the homologous DNA polymerase y or E. you DNA polymerase 1. However, neither of these putative DNA primase activities have been extensively purified or characterized. Evidence for the existence of mitochondrial DNA binding proteins was derived from studies where a human HeLa cell mitochondrial lysate was treated with formaldehyde and glutaraldehyde to achieve crosslinking of mtDNA and proteins (3). The treated DNA exhibited decreased density in CsCl gradients suggesting that crossan had occurred. Electron microscopic visualization showed mtDNA molecules with an associated membrane-like patch which upon rebanding in C80 was converted to a smaller, pronase 12 13 sensitive protein complex. The position of this protein complex was localized to specific restriction fragments known to contain the mitochondrial origin of replication. It was concluded that the HeLa cell mtDNA is attached in yilg to the inner mitochondrial membrane at or near the origin of replication and that a protein component remains associated with the mtDN A during isolation. Studies with 4,5',8-trimethylpsoralen crosslinking of [25232111113 W mt DNA demonstrated that there is likely no nucleosome structure but that the A+T rich region exhibits five uncrosslinked regions which may arise from specific protein binding (6). The presence of a nucleoprotein structure at the origin of replication was also postulated from 4, 5', 8-trimethylpsoralen crosslinking studies of the mtDNA of 121339913113 213.115 (42). These results differed in that only two uncrosslinked regions were identified, perhaps a reflection of the smaller size of the A+T region in this species (1 kb). A form of mt DNA isolated from rat liver was found to sediment more rapidly on 5- 20% sucrose gradients (>39 S) than the SDS/phenol treated DNA (39 S) suggesting that a protein component maintains the >39 S DNA in a rapidly sedimenting compact state. SDS/PAGE analysis indicated the protein component may be derived from one or more small mitochondrial inner membrane proteins (5). Further studies with this system demonstrated that the rat liver mtDN A can be isolated with a tenaciously bound polypeptide that may function to prevent branch migration of nascent strands in replicative intermediates (43). A single major polypeptide of 16,000 daltons was identified on SDS/PAGE . When crosslinked to mtDNA this polypeptide alters its density in CsCl gradients, binds selectively to the displaced single strand of the D-loop as well as to extended single- stranded regions of replicative intermediates. It has a protease (20 ug/ml proteinase K or subtilisin) insensitive remnant (6000 daltons) that remains associated with the mtDNA (44). This protein has a disproportionately high concentration of several amino acids (i.e. tyrosine and phenylalanine) suggesting that hydrophobic interactions may play an important role in protein/DNA binding (45). 14 A mitochondrial single-stranded DNA binding protein thought to be analogous to the rat liver protein has been isolated from 29.119225 lam; oocyte mitochondria. Its arrrino acid composition exhibits similarities to prokaryotic sin gle- stranded DNA binding proteins (46,47). Also, DNA binding proteins thought to have a specific affinity for supercoiled DNA (48) or double-stranded DNA encompassing the D-loop region (49) have been isolated from W laws. These DNA binding proteins may provide important structural functions or perhaps protect the DNA from nucleases, prevent branch migrational loss of nascent DNA strands or minimize secondary structtue in the displaced template strand which may impede the DNA polymerase. Presently, their exact contribution to the mechanism of mtDNA replication is not understood. WW By the early 1970's, three classes of eukaryotic cellular DNA polymerases had been identified which are denoted by a (replicative), B (repair) and y (mitochondrial, possible replicative function), in order of their discovery, to distinguish them from the prokaryotic DNA polymerases. A second potential replicative DNA polymerase was identified in 1976 and has been designated DNA polymerase 8. All DNA polymerases have three features in common. First, they utilize deoxyribonucleoside triphosphates for the synthesis of DNA by incorporation of deoxynucleoside monophosphates onto a DNA template. This requires the presence of a divalent metal ion activator and results in the subsequent release of pyrophosphate. Second, all DNA polymerases require a short DNA or RNA primer complementary to the template which provides a 3'-hydroxyl group from which DNA synthesis is initiated. Finally, DNA synthesis proceeds only in the 5'-3' direction. Replicative Enzymes - DNA polymerase 0t and DNA polymerase 8 The replicative DNA polymerase (I was first identified from calf thymus tissue (50). The ubiquitous nature of this enzyme in eukaryotic systems was established as it was subsequently purified from human HeLa cells (51), rat tissue (52-54), human lymphocytes (55), human KB cells (56), murine cells (57,58), murine myeloma cells (59), avian cells (60), rabbit (61) and calf thymus (62-64). 12132591211113 DNA polymerase or has been extensively purified and characterized (65-73). Its structural and catalytic featrnes have proven to be representative of replicative enzymes from sources as divergent as yeast (74) and human cell lines (75). DNA polymerase or is found in the soluble cytoplasmic fraction when whole cell extracts are prepared using standard methods. However, the nuclear location of the enzyme was established by use of specialized extraction procedures (76). The levels of the enzyme are thought to vary during the cell cycle, increasing timing the 61 stage before S phase (77), and are highest in rapidly growing cells. These types of studies provided evidence that DNA polymerase a is the primary cellular replicative DNA polymerase. In addition, the enzyme is required for replication of certain viral genomes, such as SV40 (7 8). The highly purified DNA polymerase (I obtained from a mum embryo homogenate was judged by SDS-polyacrylamide gel electrophoresis to be composed of four polypeptides. These had molecular weights of 148,000, 58,000, 46,000 and 43,000 daltons; the native molecular weight of the enzyme was estimated to be 280,000 (65). This DNA polymerase is completely dependent upon the presence of a DNA template, four deoxyribonucleoside triphosphates and magnesium ion for DNA polymerization. The enzyme activity is stimulated 3-fold by inclusion of 40 mM (NI-I4)2SO4 or 60 mM KCl, and is inhibited at higher salt concentrations. DNA polymerase a is most active on nicked and gapped (activated) DNA, 3-fold less active on denatured DNA and minimally active on poly (dA—dT'), while exhibiting no 15 l 6 detectable synthesis on poly (dA)-oligo (dT), poly (rA)-oligo (dT) or supercoiled Col El DNA. This form of the enzyme exhibits a high Km for dTTP (17.5 uM). It is inhibited >90% by aphidicolin at 0.3 rig/ml. No exonuclease, endonuclease, DNA-dependent or independent ATPase or RNA polymerase copurified with the DNA polymerase activity. Sedimentation of DNA polymerase a. in 10-30% glycerol gradients resulted in a single peak of DNA polymerase activity. SDS-polyacrylamide gel analysis again demonstrated that the four polypeptides could be correlated with the peak of enzymatic activity. Sedimentation under denaturing conditions indicated that DNA polymerase activity was correlated solely with the presence of the 148,000 dalton or subunit as determined by SDS-polyacrylamide gel electrophoresis (66). DNA synthesis by DNA polymerase (I on a single-stranded natural DNA template- primer was found to be dependent upon ATP or GTP (or ATP only on a poly (dT) template; 67). Synthesis was unaffected by a—amanitin at 2 [lg/ml, a concentration that inhibits RNA polymerases II and III >90%. Thus, low concentrations of DNA polymerase or were found to catalyze synthesis of short RNA primers (~12 nt) which could be extended by E. 9911 DNA polymerase 1. Using this as the basis for a primer extension assay, it was found that DNA polymerase and DNA primase activities were coincident throughout the purification scheme and suggested that DNA primase might be a part of the multisubunit DNA polymerase or. Like its prokaryotic counterpart, the dna G protein of E. 9911 (79), the DNA primase can incorporate deoxynucleotides and ribonucleotides into primers which are initiated at multiple sites on the DNA template (67). It was distinguished from the bacterial enzyme by its close association with the DNA polymerase. This close association of DNA polymerase and DNA primase was also observed in mouse (80), human KB cells (81), calf thymus (63) and in other replicative enzymes from other sources. ' The isolated polypeptides were shown to be unrelated to one another by one- dimensional peptide mapping after limited digestion with 5,. am: protease (66). Thus, 17 they are distinct enzyme subunits. Further, polyclonal antibodies against the intact DNA polymerase a and its four subunits showed that they were immunologically distinct (68) as might be predicted by their dissimilar peptide maps. These studies also showed that the or subunit (148,000 daltons) was related to higher molecular weight forms present in M suggesting that the 0: subunit was a catalytically active proteolytic fragment. To circumvent the apparent problem of proteolytic degradation a new and more rapid purification scheme was devised (69). The Km for dTTP (3.7 uM) of the new enzyme preparation was lower than that obtained with the proteolyzed form of the enzyme. All other reaction requirements and sensitivities were the same. Examination of the purified enzyme by SDS -polyacrylamide gel electrophoresis showed the presence of a polypeptide with molecular weight of 182,000 daltons, the size which was predicted from the earlier immunological studies as the undegraded form of the or subunit (68). In addition, polypeptides with molecular weights of 73,000, 60,000 and 50,000 daltons were observed. The four polypeptides were present in a ratio of 1.0/1.0/ 16] 1.2. Neither the 148,000 nor the 43,000 dalton polypeptides observed in the earlier preparation were present, but a new species of 73,000 daltons copurified with the enzyme. However, it did not bear any antigenic relatedness to the intact enzyme. The 60,000 and 50,000 dalton subunits were related antigenically to the B and 7 subunits of the original enzyme preparation (65,66). Separation of the enzyme subunits in glycerol gradients containing 2.8 M urea showed that DNA polymerase activity cosedimented with the isolated 182,000 dalton a subunit. DNA primase again copurified with DNA polymerase a and was associated with the B and/or 7 subunits (70). Template utilization studies indicated that the W DNA polymerase (1 replicates predominantly sin gle-stranded DNAs less efficiently than DN As containing short ' gaps. The rate of polymerization was ~10-fold higher on a gapped or multi-primed 18 template (71) and at 1100 nt/min approached the estimated in m rate of replication fork movement (2600 nt/min; 82). The highly purified [2195mm DNA polymerase-primase does not catalyze pyrophosphate exchange reactions, nor does it possess detectable DNA-dependent ATPase, RN ase H or 3'-5' exonuclease activities (71). The lack of 3'-5' exonuclease activity was thought to be a characteristic feature of DNA polymerase or from all sources that have been examined and was a major difference from the prokaryotic replicative enzymes (29). However, a 3'-5' exonuclease activity which may have significance for DNA replication fidelity appears to be associated with the isolated 182,000 dalton subunit of the 22929211413 enzyme (73). It is not known whether other DNA polymerase a preparations also possess this "cryptic" 3'-5' exonuclease. Two forms of DNA polymerase were identified in cytoplasmic extracts of rabbit hyperplastic bone marrow cells. The first was active on activated calf thymus DNA and was identified as DNA polymerase (I; the second enzyme is active on poly d(A-T) and, unlike DNA polymerase 0t, co-chromatographs with a 3'-5' exonuclease activity. This enzyme was designated DNA polymerase 8 and provided the first indication that there may be more than one form of eukaryotic replicative DNA polymerase (83). Subsequently, a DNA polymerase 5 from whole cell homogenates of fetal calf thymus was purified to near- homogeneity and shown to be composed of two polypeptides of 125,000 and 48,000 daltons. This enzyme is thought to consist of DNA polymerase and associated 3'-5’ exonuclease activities but is devoid of DNA primase activity (84). An auxiliary protein was identified which allows this DNA polymerase to replicate templates with low primer/template ratios (85). The auxiliary prOtein was later shown to be the proliferating cell nuclear antigen (86,87), a cell cycle regulated protein, that is also required for the in m replication of SV40 DNA (88). 19 The DNA polymerase 8 isolated from human placenta is a single, 170,000 dalton polypeptide which exhibits 3'-5' exonuclease activity (89). Like the calf enzyme described above it does not possess DNA primase activity. Calf thymus DNA polymerase or has been immunoaffinity-purified (90) and was studied in conjunction with two other forms of calf thymus DNA polymerase 8 (91) in an attempt to clarify the relationships between these enzymes. These two forms of DNA polymerase 8 (81 and 8H) differ from that described earlier in that they are thought to be multi-subunit enzymes (four or more polypeptides) with high native molecular weight (240,000 and 290,000 daltons, respectively), and possessing both DNA primase and 3'-5' exonuclease activities (92). The calf thymus DNA polymerase 81 may be related to the DNA polymerase on. It cross reacts with monoclonal antibodies specific for KB cell DNA polymerase a (92). In contrast, the DNA polymerase 8II does not exhibit any immunological cross-reactivity with the same specific antibodies. All forms of 8 polymerase are sensitive to inhibition by dideoxy T'I‘P, N-ethylmaleimide and aphidicolin (89,92). Optimal activity is obtained on synthetic templates poly d(A-T) (89) and poly (dA)-oligo (dT) (92); calf thymus DNA polymerase 811 is not able to utilize activated DNA (92). The relationship of these multiple forms of DNA polymerase 8 is unclear. Further, the relation of DNA polymerase 8 to DNA polymerase a is not known. No immunological cross-reactivity between enzymes from homologous sources has been detected. Also, they exhibit difl‘erential sensitivity to inhibitors. These results suggest that the two types of DNA polymerase may be distinct. The recent observation that the 1219552111113 DNA polymerase (1 possesses a cryptic 3'-5' exonuclease (72) adds fin-ther complexity to the situation and indicates that exonuclease activity can no longer be used as the primary basis for distinction between the replicative DNA polymerases. DNA Repair Enzymes - DNA Polymerase B This class of DNA polymerase was distinguished from the replicative enzymes on the basis of low molecular weight (S 50,000) and insensitivity to the sulfltydryl group inhibitor N-ethylmaleimide (51). DNA polymerase B has been isolated from calf thymus (93), rat liver (94,95) and many other sources although it was not originally found in unicellular eukaryotes, insects or plants (96). Since that time DNA polymerase B has been purified from protozoa (97), fungi (98-100) and plants (101-102). DNA polymerase B as isolated from calf thymus (103), human KB cells (104), mouse myeloma (105), guinea pig liver (106), rat liver (107), rat ascites hepatoma cells (108) and chick embryo (109) is composed of a single polypeptide of ~40,000 daltons and appears to be the simplest class of DNA polymerase. Peptide mapping studies have shown that the chick embryo DNA polymerase B (109) shows partial homology with the DNA polymerase B from rat ascites hepatoma cells (108). The primary structures of rat (110) and human (111,112) DNA polymerase B have been deduced from the sequences of the isolated cDNAs. Both of these enzymes are polypeptides of 335 amino acids. These and other studies indicate that the primary structure of vertebrate DNA polymerase B is highly conserved. Furthermore, DNA polymerase B shares extensive sequence similarity with terminal deoxynucleotidyltransferase (113,1 14). The levels of DNA polymerase B in the cell are low and independent of the cell cycle stage (78,115). DNA polymerase (I is found in the cytoplasm (51,116). DNA polymerase B is recovered almost entirely from isolated nuclei (117,118) and is not detectable in isolated mitochondria ( 119-122). The enzyme does not catalyze pyrophosphate exchange (104,123) and has no exonuclease activity (105). It can catalyze limited strand displacement from a nick and fill gaps to completion (124-126). Hence, it has been proposed that the enzyme may function to repair short DNA gaps (125,127). 20 2 1 Initially, several laboratories were unable to detect DNA polymerase B in We, embryos (65,97,128) and it was concluded that insects do not utilize this enzyme (97). However, the enzyme was subsequently identified and purified from Mime, embryos. Like the mammalian enzymes the Mtg DNA polymerase B is not inhibited by N-ethylmaleimide or aphidicolin; it is inhibited by dideoxy TTP, phosphate and ethidium bromide and stimulated by KCl (129). The enzyme is unusual in that its molecular weight is over twice that of the mammalian enzymes (M,=110,000). Mitochondrial Enzymes - DNA polymerase 7 DNA polymerase y, originally isolated from human HeLa cells (130), is distinguished from DNA polymerases a and B by its reaction requirements, template usage properties and sensitivity to inhibitors of DNA replication. It was originally thought to be distinct from the mitochondrial DNA polymerase that had been purified from rat liver mitochondria several years earlier (131-133). However, DNA polymerase ‘Y and the mitochondrial DNA polymerase were later shown to be identical in the chick embryo (121) and rat brain (120). Since that time the mitochondrial DNA polymerase has been partially purified from several sources: human HeLa cells (119,134), chick embryo (121,135), rat liver (119,136), rat brain (120), mouse myeloma (137) and human platelets (138). DNA polymerase 7 had not previously been purified from W or any other insect species. DNA polymerase 7 has been isolated from two cellular compartments--the nucleus and mitochondrion, while there are minor levels in the cytoplasm (120,121). While its role in the nucleus is unknown it is the only DNA polymerase present within the mitochondrion (120,121,139) and functions to replicate mtDNA. Some studies have also implicated the enzyme in the replication of the viral genomes of parvovirus H-l (140) and bovine parvovirus (141,142). Like other DNA polymerases from both prokaryotic and eukaryotic sources, DNA polymerase yrequires a primed DNA template, a divalent cation (Mg2+ or Mn2+) and the 22 four deoxyribonucleoside triphosphates to catalyze DNA synthesis. Several different laboratories observed that the mitochondrial enzyme was stimulated 5-10-fold by inclusion of high levels of monovalent salt such as KCl or NaCl (0.1-0.3 M; 119,121,134,137). These salt concentrations severely inhibit the replicative DNA polymerase (I. The mitochondrial enzyme is also stimulated by 20-50 mM potassium phosphate, concentrations which are completely inhibitory for DNA polymerase B (134). Consequently, the mitochondrial DNA polymerase is assayed under specialized conditions which allow it to be distinguished from the Other cellular DNA polymerases. The use of the synthetic template-primer poly (rA)-oligo (dT), which is not utilized by DNA polymerase 0t (65), 0.1-0.5 mM MnClz, 20-50 mM potassium phosphate (pH 8-9), 0.1- 0.3 M KCl and dTTP results in a DNA replication assay that is specific for DNA polymerase y (134). The Km for dTTP of the chick (121,135) and mouse (137) mitochondrial DNA polymerase is 1-2 M. The Km values for DNA of the mouse enzyme vary with the substrate: 13.4, 0.12 and 0.07 rig/ml (as 3'-OH) for activated calf thymus DNA, poly (rA)-oligo (dT) and poly (dA)-oligo (dT), respectively. DNA polymerase 7 activity is very sensitive to inhibition by N-ethylmaleimide (1 19-121,134,137), p-hydroxymercuribenzoate (134,138), ethidium bromide (119,121) and dideoxy TTP (29). It is not inhibited by aphidicolin (29), Ara-CTP or Ara-ATP (119). There is no inhibition of the chick embryo DNA polymerase y by anti-DNA polymerase a or anti-DNA polymerase B polyclonal immunoglobulins (121) suggesting that these three enzyme classes are antigenically and structurally unrelated. The DNA polymerase y from mouse myeloma cells is identical in chromatographic behavior and reaction requirements to that obtained from normal mouse tissue (137) and has no detectable endonuclease, exonuclease, RNase H or RNA polymerase activities. Recently, it was discovered that the chick embryo enzyme possesses a 3'-5' exonuclease 23 activity (143). This finding is very important as it may relate to the DNA replication fidelity of the enzyme. Because of its low abundance, instability and tendency to aggregate, physical characterization of the enzyme has proved to be difficult. Velocity sedimentation studies indicated that the partially purified enzyme had a sedimentation coefficient of 7-9S (120,121,134,135,l37), although one estimate for the rat liver enzyme was as low as 48 (119). Estimates of the molecular weight obtained from gel filtration studies have ranged from 180,000-300,000 daltons (121,135,137) and >240,000-270,000 daltons when estimated by nondenaturing polyacrylamide gel electrophoresis (134,137). The calculated native molecular mass values that have been reported have ranged from 60,000 daltons for the rat liver enzyme (119) to 180,000 daltons for the chick embryo enzyme (135). In most cases no assessment of the degree of purity or the subunit composition was made due to insufficient purification or lack of purified protein (119-121,137). However, SDS polyacrylamide gel electrophoresis of a human HeLa cell preparation showed two major polypeptide bands of 65,000 and 56,000 daltons, and two minor bands of 25,000 and 45,000 daltons, but no correlation with enzymatic activity was attempted (134). A similar analysis of the highly purified chick embryo enzyme revealed the presence of several polypeptides: 135,000, 110,000, 86,000 47,000 and 41,000 daltons. Only the abundance of the 47 ,000 dalton polypeptide correlated with enzymatic activity although the minimum native molecular weight of the enzyme is predicted to be 150000180000 daltons (135). Hence, it has been proposed that the chick embryo DNA polymerase ymay be a homotetramerucomposed of four identical 47 .000 dalton subunits (135). Template usage was exanrined in an effort to identify template sequence and/or structural features that affected DNA polymerase 7 activity and to compare the catalytic properties of the mitochondrial enzyme to DNA polymerases a and B. The literature indicates that in the presence of Mn2+ the various mitochondrial DNA polymerases are the most active on poly (rA)-oligo (dT) (119-121,134,137) the activity being 2-6-fold greater 24 than that obtained on poly (dA)-oligo (dT) (121,137). The activity of the human HeLa cell DNA polymerase yon poly (rA)-oligo (dT) is 20-fold less when Mg2+ is used as the divalent cation (134). In the presence of Mg2+ the activity of the mouse enzyme is 2-fold greater on poly (dA)-oligo (dT) than on poly (rA)-oligo (dT) (137). This enzyme is inactive on activated calf thymus DNA in the presence of Mn2+ (137) while the rat brain enzyme is active on that template-primer under the same conditions (121). No DNA synthetic activity is observed on native, denatured (134,137) or unprimed (120) DNAs. Further, no activity is detected on primed heteropolymeric RNA (120,121,134,137) eliminating the possibility of the presence of reverse transcriptase activity (29). The mitochondrial enzymes are inactive in the absence of template-primer suggesting that the observed DNA synthesis does not result from a terminal deoxynucleotidyltransferase activity (137). i 1' f N i Accurate DNA replication is dependent upon at least three steps that operate to reduce errors. First, there must be a mechanism to discriminate against incorporation of an incorrect nucleotide. If an incorrect nucleotide is incorporated, an exonucleolytic proofreading mechanism could allow its recognition and excision before incorporation of the next nucleotide and fixation of the error occurs. Finally, if a replication error has been made, post-synthetic repair mechanisms can operate to correct the defect. Based on a consideration of the differences in free energy of complementary versus noncomplementary base pairs, it has been predicted that the error frequency of DNA synthesis could be as high as one mispair per 10-100 nucleotides incorporated (144). However, measm'ements of spontaneous mutation frequencies indicate that the frequency of base pair substitutions is 10‘7-10‘ll per base pair replicated (145). The DNA polymerase must play an active role in accurate nucleotide polymerization, but the extent of this contribution and factors that may influence accuracy nwd to be determined in mm. 25 Early examination of DNA replication fidelity utilized assays which measured DNA polymerization on synthetic polynucleotides. The ratio of incorporation of a noncomplementary nucleotide to the complementary nucleotide was a reflection of the error rate of DNA synthesis. These studies indicated that various DNA replication proteins are active in enhancing the accuracy of DNA synthesis, and that error rates of DNA synthesis differ between the enzyme classes (146,147). However, these assays were of limited sensitivity and it was not known if the results were applicable to synthesis on natural DNA templates. Subsequently, a biological assay using a natural DNA derived from an amber mutant of bacteriophage ¢X174 was developed for measuring the error frequency by reversion of the amber mutation during in 319 DNA synthesis (Figure 4). This assay has the advantage of using natural DNA, is 2-3 orders of magnitude more sensitive than the earlier assay and allows assessment of the influence of other replication proteins on the accuracy of DNA polymerization (148). However, it is limited in that it only detects base substitution errors at l, 2 or 3 sites. These studies demonstrated that E, 9211 DNA polymerase I copied natural DNA with a high degree of accuracy. The error rate could be increased 8-fold by altering the DNA synthesis reaction conditions to include manganese in place of magnesium which is the physiologically relevant divalent cation. It was also found that other mutagenic and carcinogenic divalent cations, such as Co“, increased the error rate several fold. Further, high concentrations of incorrect nucleotide relative to correct nucleotide increased the error rate significantly, suggesting that regulation of nucleotide pool concentrations may be important to ensure DNA replication fidelity in yiyg (149,150). When the eukaryotic DNA polymerases were evaluated using the same assay system, it was found that DNA polymerases B and 7 were highly inaccurate while DNA polymerase at was 4-10-fold more accurate in DNA synthesis with a base substitution error rate of 1/50,000- 1/ 100,000 (151). Thus, it was concluded that the high degree of accuracy 26 Figure 4. Model for the ¢XI74am3 reversion assay. (Redrawn from Weymouth, L. A. and Loeb, L. A. (1978) Proc. Natl. Acad. Sci. U. S. A. 75: 1924-1928.) 27 SCHEME FOR DETERMINATION OF REPLICATION FIDELITY l. cattex'au’ 0' India lull-hut mm meat 3' \ ' II? \‘- I t W MM ox”! ex-n’ 28 observed in 1111.9 must be due to the influence of other unidentified enzyme subunits, cofactors or accessory proteins that function to regulate the accuracy of nucleotide polymerization, as well as postreplicative repair processes. Analysis of the highly purified, unproteolyzed 11259911113 DNA polymerase 0t showed that its in m DNA replication fidelity was increased 2-4—fold (71) over earlier determinations for DNA polymerase or from mouse myeloma and calf thymus (151). Further, the data suggest that the accuracy of nucleotide polymerization by M3 DNA polymerase or is nearly the same as that of E. 5:911 DNA polymerase III holoenzyme (71), an enzyme that possesses a 3'-5' exonuclease activity which has been shown to increase its in yin}: DNA replication fidelity 100-fold (152- 154). Immunoaffinity purified DNA polymerase or from calf thymus and human lymphocytes was also found to be 12-20-fold more accurate than the conventionally purified enzymes with error rates of 1/460,000-1/830,000 (155). Like the W3 enzyme, the calf thymus enzyme preparation has no detectable 3'-5' exonuclease activity (155). These results suggest that DNA polymerase or may be more efficient in nucleotide selection and incorporation than its prokaryotic counterpart since these eukaryotic enzymes did not appear to possess a proofreading exonuclease activity (71,155). However, dissociation of the 182,000 dalton subunit from the other three subunits of the 12mm DNA polymerase-primase complex unmasked a potent 3'-5' exonuclease activity (72). The ratio of DNA polymerase to 3'-5' exonuclease activity associated with the isolated 182,000 dalton subunit was ZOO-fold greater than in the intact enzyme. Further, in the ¢X174 amber reversion assay, the isolated 182,000 dalton subunit was 100-fold more accurate in nucleotide polymerization than the intact multisubunit enzyme or the isolated complex composed of only the 182,000 and 73,000 dalton subunits exhibiting an error rate of 106-10". Thus, the 3'-5' exonuclease activity of the 182,000 dalton subunit is obscured in v_itrg by the presence of the other subunits. The sole presence 29 of the 73,000 dalton subunit may be sufficient to prevent detectable 3'-5' exonuclease activity. The distinction between the DNA polymerases 0t and 8 is now less dramatic since both enzymes may possess 3'-5' exonuclease activity. However, it is not yet known if mammalian DNA polymerase-primase complexes also possess a similar cryptic 3'-5' exonuclease activity. Also, while DNA polymerase 8 has been isolated from mammalian sources (83,91) there have been no reports of the isolation of an analogous enzyme from hi1 . A forward mutational assay capable of detecting base substitution, frameshift, deletion, duplication and complex errors indicates that the differences in accuracy observed between the various classes of DNA polymerase depends upon the type of error being considered. DNA polymerase B from rat, chick embryo and human liver produce frame shift and base substitution errors with nearly equal frequency-J error per 5000 nucleotides incorporated (156). Analysis of DNA polymerase or from chick embryo, calf thymus, human KB and HeLa cells, and mouse myeloma indicate these enzymes produce single base substitution errors, but also generate single base frameshifts and, like DNA polymerase B, large deletions (157). In contrast, 90% of the errors produced by DNA polymerase y from chick embryo and calf liver are single base substitutions and while a low level of deletion mutations are observed, no frameshift errors are dewcted ( 157). Thus, when all possible sites are considered the average mutation rates for DNA polymerases B, or and y are one error per 1,500, 4,000 and 10,000 nucleotides polymerized, respectively (158). From this data it was observed that the least accurate enzyme, DNA polymerase B, was also the least processive enzyme. This means that it only incorporates one or a few nucleotides per template-primer binding event. In contrast, the most accurate enzyme, DNA polymerase'y, was also the most processive (157). This suggests that there may be a I correlation between the processivity and the fidelity of DNA replication. 111C 30 Examination of calf thymus DNA polymerase 8 (159) and chick embryo DNA polymerase y (143) demonstrated that both enzymes possess a 3'-5' exonuclease activity which fulfill all the established criteria for a proofreading exonuclease. Both of these exonuclease activities hydrolyze radiolabeled DNA in a 3'-5' direction, excise matched and mismatched primer termini with a preference for a mismatch, are inhibited by high concentrations of the deoxynucleoside triphosphates or 5' deoxynucleoside monophosphates and increase the accuracy for base substitution errors 100-fold. Even though the accuracy of DNA polymerases 'y and 8 is very high it is apparent that other as yet unidentified accessory factors, as well as DNA repair processes must play a significant role in achieving the high accuracy of DNA synthesis observed in vivg. i ' f DN 11 h i When there is a large excess of primer termini over enzyme molecules, nucleotide polymerization can be thought of as a two step, cyclic process (160). First, the enzyme binds the template-primer. Second, it catalyzes correct nucleotide selection and insertion. This step can be repeated until the enzyme dissociates fi'om the template-primer. Following dissociation the enzyme is free to bind at another primer terminus and repeat the process. The processivity of a DNA polymerase is defined as the number of nucleotides that are incorporated at a primer-terminus before dissociation of the enzyme occurs. The first efforts to quantitate the actual number of nucleotides incorporated per template-primer binding event utilized the direct measurement of the ratio of dGMP to dCMP incorporawd by Esghgjghja £911 DNA polymerase I into the right hand cohesive end of bacteriophage phage A DNA which contains the sequence 5'-G-G-G-C-G-G-C-G-3'. These studies showed that the enzyme incorporated more than one nucleotide per enzyme binding event, but the method was limited by the length of the template (161). Subsequently, the processivity of Escherichia 9.911 DNA polymerase I was quantitatively determined by a method which allows comparison of the rates of polymerization in the 31 presence of a limited and a complete complement of the four deoxynucleoside triphosphates. These studies confirmed the moderately processive nature of the prokaryotic enzyme on natural DNAs (2050 nucleotides) and showed that template structure, reaction temperature and ionic strength all affect processivity. Further, they demonstrated the distinctly non-processive nature of DNA synthesis of the human KB cell DNA polymerase B (160). This kinetic method was also used to assess the processivity of Eth'chia 9911 DNA polymerase 111 core and holoenzyme forms on primed bacteriophage fd DNA. This provided the first evidence that particular enzyme subunits present in the holoenzyme result in increased processivity over that observed with the core enzyme (95 vs. 12 nucleotides), and also showed that the processive synthesis of the holoenzyme is stimulated 2-fold by the Eghefichia coll single-stranded DNA binding protein (162). More direct methods of the assessment of processivity confirmed the results obtained by the quantitative kinetic method. These involved fractionation of synthetic DNA products by gel filtration (162) or by denaturing polyacrylamide gel electrophoresis. These studies demonstrated that the presence of the B subunit is responsible for the highly processive DNA synthesis of Escherichia 9911, DNA polymerase 11] holoenzyme (163). Denaturing gel analysis demonstrated that the processivity on singly-primed M13 DNA of the m DNA polymerase-primase, its 182,000 dalton subunit and the combined 182,000/7 3,000 dalton subunits are all identical with 20-30 nucleotides incorporated per primer-terminus (72). This study also showed that only the processivity of the isolated 182,000 dalton subunit is increased 2-3-fold by inclusion of W 9211 single-stranded DNA binding protein (S SB). The processivity of the other enzyme assemblies is not increased by SSB. The chick embryo and mouse DNA polymerase 0t synthesize short DNA segments \3 050 nucleotides) as determined by alkaline sucrose gradient centrifugation and denatlning gel analysis of the products of processive synthesis, respectively (164,165). 32 Thus, it has been concluded that the eukaryotic nuclear DNA polymerases are only moderately processive. This result was unexpected as several well studied prokaryotic DNA polymerases such as We £8 DNA polymerase III holoenzyme (163), bacteriOphage T4 DNA polymerase (166) and T7 DNA polymerase (167) are known to polymerize thousands of nucleotides before dissociation from the template. Recently however, it was demonstrated that the calf thymus DNA polymerases a and 8, which were originally thought to have low processivity values (91), are capable of highly processive synthesis when the reaction pH and magnesium concentration are decreased below the optimal values required for maximum synthetic rate (168). Also, another form of calf thymus DNA polymerase 8 (84) appears to be stimulated by an auxiliary protein in a manner analogous to the B subunit of Escherichia 9011 DNA polymerase III holoenzyme (85). This auxiliary protein has been shown to be the proliferating cell nuclear antigen (86,87), a cell cycle regulated protein that is also required for SV40 DNA replication (88). The chick embryo DNA polymerase B is non-processive on a synthetic template- primer (164) in agreement with the original result obtained with the human KB cell DNA polymerase B (160). This is in contrast to DNA polymerase yfrom the same source which is proposed to be a highly processive enzyme on a synthetic ribohomopolymer (169). It is clear that the question of eukaryotic DNA polymerase processivity is complex. It is likely to be influenced by many factors in m which are not well defined. Further, the more complex question of what accessory proteins, factors and environmental conditions are relevant for high processivity in $3.9 remains unanswered. 10. 11. 12. 13. 14. 15. 16. 17. 18. REFERENCES Clayton, D. A. (1982) Cell 28: 693-705. Bogenhagen, D. and Clayton, D. A. (1977) Cell 11: 719-727. Albring, M., Griffith, J. and Attardi, G. (1977) Proc. Natl. Acad. Sci. USA. 74: 1348- 1352. Pinon, H., Barat, M., Tourte, M., Dufresne, C. and Mounolou, J. C. (1978) Chromosoma 65: 383-389. VanTuyle, B. C. and McPherson, M. L. (1979) J. Biol. Chem. 254: 6044-6053. Potter, D. A., Fostel, J. M., Beminger, M., Pardue, M. L. and Cech, T. (1980) Proc. Natl. Acad. Sci. USA. 77: 4118-4122. Fauron, C. M.-R. and Wolstenholme, D. R. (1976) Proc. Natl. Acad. Sci. USA. 73: 3623-3627. Polan, M. L., Friedman, S., Gall, J. F. and Gehring, W. (1973) J. Cell. Biol. 56: 580-589. Bultrnann, H. and Laird, C. D. (1973) Biochem Biophys. Acta . 299: 196-209. Peacock, W. J ., Brutlag, D., Goldring, E., Appels, R., Hinton, C. and Lindsley, D. C. (1974) Cold Spring Harbor Symp. Quant. Biol. 38: 405-415. Klukas, C. K. and Dawid, I. B. (1976) Cell 9: 615-625. Goldring, E. S. and Peacock, W. J. (1977) J. Cell. Biol. 73: 279-286. Fauron, C. M.-R. and Wolstenholme, D. R. (1977) J. Cell. Biol. 75: 3113. Goddard, J. M. and Wolstenholme, D. R. (1980) Nuc. Acids Res. 8: 741-757. Christiansen, C.., Christiansen, G. and Bak, A. L. (1974) J. Mol. Biol. 84: 65- 82. Goddard, J. M. and Wolstenholme, D. R. (1978) Proc. Natl. Acad. Sci. USA . 75: 3886-3890. 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. Bonner, J. J., Beminger, M. and Pardue, M. L. (1977).Cold Spring Harbor Symp. Quant. Biol. 42: 803-814. 33 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 34 Clary, D 0., Goddard, J. M., Martin, S. M., Fauron, C. M.-R., and Wolstenholme, D. R. (1982) Nucl. Acids Res. 10: 6619-6637. Anderson, S., Bankier, A. T., Barrel], B. G., de Bruijn, M. H. 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, I. G. (1981) Nature 290: 457- 465. Bibb, M. J ., Van Etten, R. A., Wright, C. T., Walberg, M. W .and Clayton, D. A. (1981) Cell 26: 167-180. Wolstenholme, D. R. and Dawid, I. B. (1968) J. Cell. Biol. 39: 222-228. Attardi, G., Crews, S., Nishiguchi, J., Ojala, D. and Pasakony, J. (1977) Cold Spring Harbor Symp. Quant. Biol. 43: 179-192. Koike, K., Kobayashi, M. and Sekiya, T. (1978) Cold Spring Harbor Symp. Quant. Biol. 43: 193-201. Anderson, S., DeBruijn, M. H. L., Coulson, A. R., Eperon, I. C., Sanger, F. and Young, 1. G. (1982) J. Mol. Biol. 156: 683-717. Montoya, J ., Ojala, D. and Attardi, G. (1981) Nature 290: 465-470. 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. Kornberg, A. (1980) DNA Replication, W. H. Freeman and Co., San Francisco. McMacken, R., Silver, L. and Georgopoulos, C. (1986) DNA Replication In: Escherichia call and Schmidt: maintain Cellular and Molecular Biology, Vol. 1., Ingraham, J., Low, K. B., Magasanik, B., Schaechter, M., Umbarger, H. E. and Neidhardt, F. C. (eds.) pp. 564-612. Robberson, D. L., Kasamatsu, H. and Vinograd, J. (1972) Proc. Natl. Acad. Sci. USA. 69: 737-741. Robberson, D. L. and Clayton, D. A. (1972) Proc. Natl. Acad. Sci. USA. 69: 3810-3814. Kasamatsu, H. and Vinograd, J. ( 1973) Nature New Biol. 241: 103-105. Brown, W. M. and Vinograd, J. (1974) Proc. Natl. Acad. Sci. USA . 71: 4617- 4621. Robberson, D. L., Clayton, D. A. and Morrow, J. F. (1974) Proc. Natl. Acad. Sci. USA. 71: 4447-4451. Kasamatsu, H., Robberson, D. L. and Vinograd, J. (1971) Proc. Natl. Acad. Sci. U. S. A. 68: 2252-2257. ' Wong, T. W. and Clayton, D A. (1985) J. Biol. Chem. 260: 11530-11535. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 5'7 - 35 Rubenstein, J. L. R., Brutlag, D. and Clayton, D. A. (1977) Cell 12: 471-482. Lee, C. S., Davis, R. and Davidson, N. (1970) J. Mol. Biol. 48: 1-22. Chang, D. D., Hauswirth, W. W. and Clayton, D. A. (1985) The EMBO Journal 4: 1559-1567. Ledwith, B. J., Manam, S. and VanTuyle, G. C. ( 1986) J. Biol. Chem. 261: 6571-6577. Pardue, M. L., Fostel., J. M. and Cech, T. R. (1984) Nucl. Acids Res. 12: 1991- 1999. VanTuyle, G. C. and Pavco, P. A. (1981) J. Biol. Chem. 256: 12772-12779. VanTuyle, G. A. and Pavco, P. A. (1985) J. Cell. Biol. 100: 251-257. Pavco, P. A. and VanTuyle, G. A. (1985) J. Cell. Biol. 100: 258-264. Mignotte, B., Barat, M., Marsault, J. and Mounolou, J .-C. (1983) Biochem. Biophys. Res. Comm. 117: 99-107. Mignotte, B., Barat, M. and Mounolou, J .-C. (1985) Nucl. Acids Res. 13: 1703- 17 16. Mignotte, B. and Barat, M. (1986) Nucl. Acids Res. 14: 5969-5980. Cordonnier, A. M., Dunon-Bluteau, D. and Brun, G. (1987) Nucl. Acids Res. 15: 477-490. Bollum, F. J. (1960) J. Biol. Chem. 235: 2399-2403. Weissbach, A., Schlabach, A., Fridlender, B., and Bolden, A. (1971) Nature New Biol. 231: 167-170. Baril, E. F., Brown, 0. E., Jenkins, M. D., and Laszlo, J. (1971) Biochemistry 10: 981-1992. DeRecondo, A. M., Lepesant, J. A., Fichot, 0., Grasset, L., Rossignol, J. M., and Cazillis, M. (1973) J. Biol. Chem. 248: 131-137. Holmes, A.M., Hesslewood, I. P., and Johnston, 1. R. (1974) Eur. J. Biochem. 43: 487-499). Smith, R. G. and Gallo, R. C. (1972) Proc. Natl. Acad. Sci. U. S. A . 69: 2879- 2884. Sedwick, W. D., Wang, T. S. F., and Korn, D. (1972) J. Biol. Chem. 247: 5026-5033. Adams, R. L. P., Henderson, M. A. L., Wood, W., and Lindsay, J. G. (1973) Biochem. J. 131: 237-246. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75- '16- 36 Yamaguchi, M., Hendrickson, E. A. and dePamphilis, M. L. (1985) J. Biol. Chem. 260: 6254-6263. Persico, F.J., Nicholson, D. E., and Gottlieb, A. A. (1973) Cancer Res . 33: 1210-1216. ' Brun, G., Rougeon, F., Lauber, M., and Chappeville, F. (1974). Eur J. Biochem. 41: 241-251. Eggnes, J. J., Downey, K. M. and So, A. G. (1973) Biochemistry 12: 4378- 4. Grosse, F. and Krauss, G. (1981) Biochemistry 20: 5470-5475. Chang, L. M. S., Rafter, E., Augl, C. and Bollum, F. J. (1984) J. Biol. Chem. 259: 14679-14687. Wahl, A. F., Kowalski, S. P., Harwell, L. W., Lord, E. M. and Bambara, R. A. (1984) Biochemistry 23: 1895-1899. Banks, G. R., Boezi, J. A. and Lehman, I. R. (1979) J. Biol. Chem. 254: 9886- 9892. Villani, G., Sauer, B. and Lehman, I. R. (1980) J. Biol. Chem. 255: 9479-9483. Conaway, R. C and Lehman, I. R. (1982) Proc. Natl. Acad. Sci. USA. 79: 2523-2527. Sauer, B. and Lehman, I. R. (1982) J. Biol. Chem. 257: 12394-12398. Kaguni, L. S., Rossignol, J.-M., Conaway, R. C. and Lehman, I. R. (1983) Proc. Natl. Acad. Sci. USA. 80: 2221-2225. Kaguni, L. S., Rossignol, J. M., Conaway, R. C., Banks, G. R. and Lehman, I. R. (1983) J. Biol. Chem. 258: 9037-9039. Kaguni, L. S., DiFrancesco, R. A. and Lehman, I. R. (1984) J. Biol. Chem. 259: 9314-9319. Cotterill, S. M., Reyland, M. B., Loeb, L. A. and Lehman, I. R. (1987) Proc. Natl. Acad. Sci. USA. 84: 5635-5639. Cotterill, S., Chui, G. and Lehman, I. R. (1987) J. Biol. Chem. 262: 16100- 16104. Plevani, P., Foiani, M., Valsasnini, P., Badaracco, G., Cheriathundarn, E. and Chang, L. M. S. (1985) J. Biol. Chem. 260: 7102-7107. Wong, W. W., Paborsky, L. R., Fisher, P. A., Wang, T. S.-F. and Kom, D. (1986) J. Biol. Chem. 261: 7958-7968. Lynch, W. E., Surrey, S. and Lieberman, 1. (1975) J. Biol. Chem. 250: 8179- 8183. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 37 Chiu, R. W. and Baril, E. F. (1975) J. Biol. Chem. 250: 7951-7957. Li, J. J. and Kelly, T. J. (1984) Proc. Natl. Acad. Sci. U. S. A. 81: 6973-6977. Rowen, L. and Kornberg, A. (1978) J. Biol. Chem. 253: 770-774. Yagura, T., Tanaka, S., Kozu, T., Seno, T. and Kom, D. (1984) J. Biol. Chem. 258: 6698-6700. Hu, S.-Z., Wang, T. S.-F. and Korn, D. (1984) J. Biol. Chem. 259: 2602-2609. Blumenthal, A. B., Kriegstein, H. J. and Hogness, D. S. (1973) Cold Spring Harbor Symp. Quant. Biol. 38: 205-223. Bymes, J. J., Downey, K. M., Black, V. L. and So, A G. (1976) Biochemistry 15: 2817-2823. Lee, M. Y. W. T., Tan, C.-K., Downey, K. M. and 80, A. G. (1984) Biochemistry 23: 1906-1913. Tan, C.-K., Castillo, C, S0, A. G. and Downey, K. M. (1986) J. Biol. Chem. 261: 12310-12316. Bravo, R., Frank, R., Blundell, P. A. and MacDonald-Bravo, H. (1987) Nature 326: 515-517. Prelich, G., Tan, C.-K., Kostura, M., Mathews, M. B., 80, A. G., Downey, K. M. and Stillman, B. (1987) Nature 326: 517-520. Prelich, G., Kostura, M., Marshak, R., Mathews, M. B. and Stillman, B. (1987) Nature 326': 471-475. Lee, M. Y. W. T. and Toomey, N L. (1987) Biochemistry 26: 1076-1085. Wahl, A. F., Kowalski, S. P., Harwell, L. W., Lord, E. M. and Bambara, R. A. (1984) Biochemistry 23: 1895-1899. Crute, J. J ., Wahl, ‘A. F. and Bambara, R. A. (1986) Biochemistry 25: 26-36. Wahl, A. F., Crute, J. J., Sabatino, R. D., Bodner, J. B., Marraccino, R. L., Harwell, L. W., Lord, E. M. and Bambara, R. A. (1986) Biochemistry 25: 7821- 7827. Chang, L. M. S. and Bollum, F. J. (1971) J. Biol. Chem. 246: 5835-5837. Haines, M. E., Holmes, A. M., and Johnston, 1. R. (1971) FEBS Letts 17: 63- 67. Berger, Jr., H., Huang, R. C. and Irwin, J. L. (1971) J. Biol. Chem. 246: 7275- 7283. Chang, L. M. S. (1976) Science 191: 1183-1185. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 1 15. \16. 38 Fumkawa, Y., Yamada, R. and Kohno, M. ( 1979) Nucl. Acids Res. 7: 2387- 2398. Baril, E. F., Scheiner, C. and Pederson, T. (1980) Proc. Natl. Acad. Sci. USA. 77: 3317-3321. Schiebel, W. and Rafael, A. (1980) FEBS Lett. 121: 81-85. Sakaguchi, K. and Lu, B. C. (1982) Mol. Cell. Biol. 2: 752-757. Yamaguchi, M., Chou, M.-Y., Matsumoto, M. and Fuasawa, H. (1979) Jpn. J. Genet. 54: 97-108. Sakaguchi, K., Hotta, Y. and Stern, H. (1980) Cell Struct. Funcr. 5: 323-324. Chang, L. M. S. (1973) J. Biol. Chem. 248: 3789-3795. Wang, T. S. F., Sedwick, W. D. and Korn, D. (1975) J. Biol. Chem. 250: 7040- 7044. Tanabe, K., Bohn, E. W., and Wilson, S. H. (1979) Biochemistry 18: 3401- 3406. Kunkel, T. A., Tcheng, J. E. and Meyer, R. R. (1978) Biochim. Biophys. Acta. 520: 302-316. Stalker, D. M., Mosbaugh, D. W. and Meyer, R. R. (1976) Biochemistry 15: 31 14-3121. Ono, K., Ohashi, A., Tanabe, K., Matsukage, A., Nishizawa, M. and Takahashi, T. (1979) Nucl. Acids Res. 7: 715-726. Yamaguchi, M., Tanabc, K. Taguchi, Y., Nishizawa, M., Takahashi, T. and Matusukage, A. (1980) J. Biol. Chem. 255: 9942-9948. Zmudzka, B. Z., SenGupta, D., Matsukage, A., Cobianchi, F., Kumar, P. and Wilson, S. H. (1986) Proc. Natl. Acad. Sci. USA. 83: 5106-5110. SenGupta, D. N.., Zmudzka, B. Z., Kumar, R, Cobianchi, F., Skowronski, J. and Wilson, S. H. (1986) Biochem. Biophys. Res. Comm. 136: 341-347. Abbotts, J., SenGupta, D. N., Zmudzka, B., Widen, S. G., Notario, V. and Wilson, 8. H. (1988) Biochemistry 27: 901-909.. Matsukage, A., Nishikawa, K. , Ooi, T. Seto, Y.and Yamaguchi, M.(1987) J. Biol. Chem. 262: 8960-8962. Anderson, R. S., Lawrence, C. B., Wilson, S. H. and Beattie, K. L. (1987) Gene 60: 163-173. Chang, L. M. S. , Brown, M. and Bollum, R. J. (1983) J. Mol. Biol. 74: 1-8. Weissbach, A. (1975) Cell 5: 101-108. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 39 Furia, M.,Polito, L. C., Locorotondo, G. and Grippo, P. (1979) Nucl. Acids Res. 6: 3399-3410. Hiibscher, U, Kuenzle, C. C. and Spadari, S. (1979) Proc. Natl. Acad. Sci. USA. 76: 2316-2320. Bolden, A., Pedrali Noy, G. and Weissbach, A. (1977) J. Biol. Chem. 252: 3351- 3356. Hugscher, U., Kuenzle, C. C. and Spadari, S. (1977) Eur. J. Biochem. 81: 249- 25 . Bertazzoni, U., Scovassi, A. I. and Brun, G. M. (1977) Eur. J.. Biochem. 81: 237-248. Scovassi, A. I., Wicker, R. and Bertazzoni, U. (1979) Eur. J. Biochem. 100: 491-496. ' Chang, L. M. S.and Bollum, F. J. (1973) J. Biol. Chem. 248: 3398-3404. Wang, T. S.-F. and Korn, D. (1980) Biochemistry 19: 1782-1790. Mosbaugh, D. W. and Linn, S. (1984) J. Biol. Chem. 259: 10247-10251. Friedberg, E. C. (1985) DNA Repair, pp. 251-357, Freeman Publications, New York. Bertazzoni, U., Stefanini, M., Pedrali Noy, G., Biulotto, B., Nuzzo, F., Falaschi, A. and Spadari, S. (1976) Proc. Natl. Acad. Sci. U.S.A. 73: 785-789. Dusenbery, R. L. and Smith, P. D. (1983) Dev. Genett. 3: 309-327. Sakaguchi, K and Boyd, J. B. (1985) J. Biol. Chem. 260: 10406-10411. Fridlender, B., Fry, M., Bolden, A., and Weissbach, A. (1972) Proc. Natl. Acad. Sci. U.S.A . 69: 452-455. Kalf, C. F. and Ch'ih, J. J. (1968) J. Biol. Chem. 243: 4904-4916. Meyer, R. R. and Simpson, M. V. (1968) Proc. Natl. Acad. Sci. USA. 61:130- 137. Meyer, R. R. and Simpson, M. v (1970) J. Biol. Chem. 245: 3426-3435. Knopf, K. W., Yamada, M. and Weissbach, A. (1976) Biochemistry 15: 4540- 4548. Yamaguchi, M., Matsukage, A. and Takahashi, T. (1980) J. Biol. Chem. 255: 7002-7009. Tanaka, S. and Koike, K. (1977) Biochem. Biophys. Acta. 479: 290-299. Matsukage, A., Bohn, E. W. and Wilson, S. H. (1975) Biochemistry 14: 1006- 1020. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 40 Tarrago-Litvak, L., Desgranges, C., Araya, A and Litvak, S. (1979) Eur. J. Biochem. 93: 271-278. Geuskens, M., Hardt, M., Pedrali Noy, G. and Spadari, S. (1981) Nucl. Acids Res. 9: 1599-1613. Kollek, R. and Goulian, M. (1981) Proc. Natl. Acad. Sci. USA. 78: 6206- 6210. Robertson, A. T., Bates, R. C. and Stout, E. R. (1983) Biochem. Biophys. Res. Comm. 117: 580-586. Fry, M. and Loeb, L. A. (1986) Animal Cell DNA Polymerases, CRC Press, Boca Raton. Kunkel, T. A. and Soni, A. (1988) J. Biol. Chem. 263: 4450-4459. Loeb, L. A. and Kunkel, T. A. (1982) Ann. Rev. Biochem. 52: 429-457. Drake, J. W. (1969) Nature 221: 1132. Krauss S. W. and Linn, S. (1980) Biochemistry 19: 220-228. Krauss, S. W. and Linn, S. (1982) Biochemistry 21: 1002-1009. Weymouth, L. A. and Loeb, L. A. (1978) Proc. Natl. Acad. Sci. U.S.A. 75: 1924-1928. Kunkel, T. A. and Loeb, L. A. (1979) J. Biol. Chem. 254: 5718-5725. Kunkel, T. A. and Loeb, L. A. (1980) J. Biol. Chem. 255: 9961-9966. Kunkel, T. A. and Loeb, L. A. (1981) Science 213: 765-767. Kunkel, T. A., Schaaper, R. M., Beckman, R. A. and Loeb, L. A. (1981) J. Biol. Chem. 256: 9883-9889. Fersht, A. R., Knill-Jones, J. W. and Tsui, W.-C. (1982) J. Mol. Biol . 156: 37- 51. Echols, H., Lu, C. and Burgers, P. M. J. (1983) Proc. Natl. Acad. Sci. U.S.A. 80: 2189-2192. Reyland, M. E. and Loeb, L. A. (1987) J. Biol. Chem. 262: 10824-10830. Kunkel, T. A. ( 1985) J. Biol. Chem. 260: 5787-5796. Kunkel, T. A. (1985) J. Biol. Chem. 260: 12866-12874. Kunkel, T. A. and Alexander, P. S. (1986) J. Biol. Chem. 261: 160-166. Kunkel, T. A., Sabatino, R. D. and Bambara, R. A. (1987) Proc. Natl. ACad. Sci. U.S.A. 84: 4865-4869. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 41 Bambara, R. A., Uyemura, D. and Choi, T. (1978) J. Biol. Chem. 253: 413-423. Uyemura, D., Bambara, R. and Lehman, I. R. (1975) J. Biol. Chem. 250: 8577- 8584. Fay, P. J., Johanson, K. 0., McHenry, C. S. and Bambara, R. A. (1981) J. Biol. Chem. 256: 976-983. Crute, J. J., LaDuca, R. J., Johanson, K. 0., McHenry, C. S. and Bambara, R. A. (1983) J. Biol. Chem. 258: 11344-11349. Matsukage, A., Yamaguchi, M., Tanabe, K., Taguchi, Y. N., Nishizawa, M. and Takahashi, T. (1983) "New Approaches in Eukaryotic DNA Replication", A. M. deRecondo (ed.), pp. 81-106, Plenum Press, New York. Detera, S. D., Becerra, S. P., Swack, J. A. and Wilson, 8. H. (1981) J. Biol. Chem. 256: 6933-6943. Mace, D. C. and Alberts, B. M. (1984) J. Mol. Biol. 177: 313-327. Tabor S. and Richardson, C. C. (1987) Proc. Natl. Acad. Sci. U.S.A. 84: 4767- 4771. Sabatino, R. D., Myers, T. W., Bambara, R. A., Kwon-Shin, 0., Marraccino, R. L. and Frickey, P. H. (1988) Biochemistry 27: 2998-3004. Yamaguchi, M., Matsukage, A. and Takahashi, T. (1980) Nature 285: 45-47. Chapter II A Mitochondrial DNA Polymerase from Embryos of Dr hi] 1 t r: Purification, Subunit Structure and Partial Characterization 42 ABM The mitochondrial DNA polymerase has been purified to near homogeneity from early embryos of Drgsgphila W. Sodium dodecyl sulfate gel electrophoresis of the highly purified enzyme reveals two polypeptides with Mrs of 125,000 and 35,000, in a ratio of 1:1. The enzyme has a sedimentation coefficient of 7.6 S and a Stokes radius of 51 A. Taken together, the data suggest that the D. W DNA polymerase 7 is a heterodimer. DNA polymerase activity gel analysis has allowed the assignment of the DNA polymerization function to the large subunit. The DNA polymerase exhibits a remarkable ability to utilize efficiently a variety of template-primers including gapped DNA, poly r(A)-oligo d(T) and singly-primed ¢Xl74 DNA. Both the crude and the highly purified enzymes are stimulated by KC], and inhibited by dideoxythymidine triphosphate and by N-ethylmaleimide. Thus, the catalytic properties of the near-homogeneous mm enzyme are consistent with those of DNA polymerase y as partially purified from several vertebrates. 43 W Three major classes of DNA polymerases have been isolated from eukaryotes, and are designated as replicative, repair and mitochondrial enzymes. The replicative DNA polymerase at is a nuclear enzyme which functions in chromosome replication (1). A high molecular weight multi-subunit form of this DNA polymerase was pm'ified from 21959121133 melanogaster; embryos, and was shown to contain both a DNA polymerase and a DNA primase activity (2). DNA polymerase B, also located in the nucleus, is associated with DNA repair processes (1). DNA polymerase B as purified from a variety of sources is a low molecular weight single-subunit DNA polymerase of ~40,000 daltons (1). While investigators in several laboratories were unable to demonstrate the presence of DNA polymerase B in W embryos (3-5), a recent report has described its purification to near- homogeneity : the Mil—11.11% DNA polymerase B appears to consist of a single 110,000 dalton polypeptide (6). Because its activity is not inhibited by the sulfhydryl group blocking agent N -ethylmaleimide (NEM), it is distinguished from DNA polymerase 0t. The mitochondrial DNA polymerase is poorly characterized, perhaps because it is the least abundant enzyme: DNA polymerase 7 accounts for only about one-percent of the cellular DNA polymerase activity. Though DNA polymerase 7 has been purified partially from several sources (7-10), it has been purified highly only from a single source--chick embryos (11). DNA polymerase y is distinguished from DNA polymerases a and B by its inhibition both by NEM and by the nucleotide analog dideoxythymidine triphosphate (d2’ITP, 1). Our laboratory has begun an analysis of the biochemical and genetic requirements for mitochondrial DNA replication in Dmsgphfla embryos. Initial efforts have been directed at the idendfication, purification, and characterization of DNA polymerase y from embryonic mitochondria. This chapter describes the purification of the mitochondrial DNA 44 45 polymerase to near-homogeneity and the examination of its subunit structure and catalytic properties. LPR ED Materials Nucleotides and Nucleic Acids - Unlabeled deoxy- and ribonucleoside triphosphates were purchased from P-L Biochemicals; [3H]dTI'P, 0t[32P]dTTP and 0t[32P]dCTP were from New England Nuclear. 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 (12). Poly (rA).p(dT)10 and poly (dA).p(dT)10 were purchased from P-L Biochemicals, and contained adenine and thymine in a molar ratio of 20:1, such that the average single-stranded DNA region between primers was 200 nucleotides. Singly-primed ¢X174 DNA was as described (13). Chromatography and Buffers - Phosphocellulose P11 was purchased from Whatman; octyl-Sepharose and Sephacryl S-200 were from Pharmacia. Single-stranded DNA cellulose was prepared as described (14). Cibacron Blue-agarose was a gift of Dr. Clarence Suelter of this department. Enzymes and Protein Standards - Mia DNA polymerase a (fraction VI, 7122 u/mg) was prepared as described (2). E, 9911 DNA polymerase I (DNA polymerase I) was purchased from New England Biolabs. E. 9911 RNA polymerase was the gift of Dr. Jon Kaguni of this department. Rabbit muscle pyruvate kinase, bovine carbonic anhydrase and E. col; alkaline phosphatase were from Worthington. Bovine serum albumin (fraction V), bovine liver catalase, rabbit muscle glycogen phosphorylase b, rabbit muscle L-lactate dehydrogenase, yeast alcohol dehydrogenase, E. 9911 B-galactosidase and cytochrome c were purchased from Sigma. 46 Chemicals - PMSF (Sigma) was prepared as a 0.1 M stock solution in isopropanol and kept frozen at -20°C. Sodium metabisulfite (Baker) was prepared as a 1.0 M stock solution at pH 7.5 and stored at -20°C. Leupeptin purchased from the Peptide Institute, Minoh-Shi, Japan was prepared as a 1 mg/ml stock solution in 0.1 M potassium phosphate buffer, pH 7 .5 and stored at 4°C. Ammonium sulfate and sucrose, both ultra-pure, were from Schwarz-Mann. D'IT, Triton X-100, acrylamide, soluble potato starch, potassium iodide, and dimethyldichlorosilane were purchased from Sigma. N-ethylmaleimide (NEM) was purchased from Sigma and dissolved in water immediately before use. Cholic acid (Sigma) was dissolved in hot ethanol, filtered through Norit A (Baker) and recrystallized twice before titration to pH 7.4 with sodium hydroxide. Sodium dodecyl sulfate (SDS) for general gel electrophoresis was from Pierce; SDS for activity gel analysis was the gift of Dr. Geoffrey Banks from the National Institute for Medical Research, London, England. Collodion membranes and nitrocellulose filters were fi'om Schleicher and Schuell. Methods Processing of MB Embryos - 12. W (Oregon R) embryos of average age, 9 h were collected immediately before use and washed and dechorionated as described (15). Preparation of Partially Purified Mitochondria - The processed embryos were suspended at a ratio of 4 ml/gram wet weight in homogenization buffer (HB) containing 15 mM HEPES, pH 8, 5 mM KCl, 2 mM CaClz, 0.5 mM EDTA, 0.5 mM DTT, 0.27 M ultra-pure sucrose, 1 mM PMSF, 10 mM sodium metabisulfite and 2 ug/ml leupeptin and homogenized in 40 ml portions by three strokes of a stainless steel-Teflon homogenizer. The homogenate was filtered through a 75 um Nitex screen; the retentate was rehomogenized in the same buffer (1 Wm), filtered as above, and combined with the original filtrate. The combined filtrate was centrifuged at 1000 x g for 7 min at 3°C. The 47 supernatant fluid was recovered and the centrifugation procedure was repeated twice. The resulting supernatant fluid was centrifuged at 12,000 x g for 10 min at 3°C. The mitochondrial pellet was resuspended at a ratio of 2 ml HB/gram of starting embryos and the centrifugation was repeated. The second pellet was resuspended at a ratio of 2 mllgram and centrifuged at 12,000 x g for 15 min at 3°C. The final mitochondrial pellet was frozen in liquid nitrogen and stored at -80°C. Preparation of the Mitochondrial Extract - Frozen, partially purified mitochondria from freshly harvested and dechorionated Ma embryos (200 g) were thawed on ice for 30 min, and then suspended at a ratio of 0.5 ml/gram of starting embryos in 25 mM HEPES, pH 8.0, 10% glycerol, 0.3 M NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF, 10 mM sodium metabisulfite, and 2 ug/ml leupeptin. Sodium cholate was added to a final concentration of 2% and the suspension was incubated on ice for 30 min with mixing by inversion at 5 min intervals. The resulting extract was centrifuged at 96,000 x g for 30 min at 3°C. The supernatant fluid was recovered and an equal volume of buffer containing 25 mM HEPES, pH 8.0/ 2 mM EDTA/ 80% glycerol was added. The mitochondrial extract (fraction 1) was stored at -20°C. DNA Polymerase Assay - Reaction mixtures (0.025ml) contained 50 mM Tris-HCl, pH 8.5, 200 mM KCl, 5 mM MgC12, 5 mM 2-mercaptoethanol, 400 ug/ml BSA, 60 BM each of dATP, dCI'P, and dGTP, 30 BM [3H]d'l'I'P (2800 cpm/pmol), 250 [lg/ml activated calf thymus DNA, and enzyme. Incubation was for 20 min at 30°C. One unit of activity is that amount that catalyzes the incorporation of 1 nmol of dNTP into acid- insoluble material in 60 min at 30°C. Purification - All potassium phosphate buffers were at pH 7.6 and contained 2 mM dithiothreitol (D11), 2 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 mM sodium metabisulfite, leupeptin at 2 ug/ml and 20% glycerol. Where indicated, buffers contained 9% glycerol and/or 0.015% Triton X- 100. All operations were performed at 0- 4°C. The ionic strength of buffers was determined using a Radiometer conductivity meter. 48 Phosphocellulose Chromatography and Ammonium Sulfate Fractionation - Fraction 1 was diluted to an ionic equivalent of 70-80 mM potassium phosphate and loaded onto a phosphocellulose column (6.5 x 4 cm; 6 mg protein per packed ml phosphocellulose) equilibrated with 80 mM potassium phosphate buffer at a flow rate of 100 ml/hour. The column was washed with 375 ml of 100 mM potassium phosphate buffer at a rate of 250 ml/horn' and then a 375 ml linear gradient from 150 to 350 mM potassium phosphate was applied. The DNA polymerase activity eluted at 200 mM potassium phosphate (Figure 1A). Active fractions were pooled (fraction II) and adjusted with 80% sucrose to a final concentration of 10%. After addition of 1.2 volumes of saturated ammonium sulfate, pH 7.5 to achieve 55% of saturation at 0°C, the suspension was incubated on ice for 2 hours. The precipitate was collected by centrifugation at 96,000 x g for 30 min at 3°C, resuspended in 2.0 ml of 10 mM potassium phosphate buffer containing 45% glycerol, and stored at -20°C (fraction IIb). DNA-Cellulose Chromatography - Fraction [lb was dialyzed against 10 mM potassium phosphate buffer in a collodion bag (Mr cutoff, 25 ,000) until an ionic equivalent of 60-70 mM KCl was reached. It was then loaded onto a single-stranded DNA-cellulose column (2.8 x 3.2 cm; 0.6 mg protein per packed ml DNA-cellulose) equilibrated with 20 mM potassium phosphate buffer containing 10% sucrose at a flow rate of 10 ml/hour. At this step in the purification and for all subsequent steps, all laboratory ware (which was mosfly polypropylene) was silanized as described by Maniatis et a1. (23) to prevent enzyme loss by adsorption. The column was washed with 40 ml of 20 mM potassium phosphate buffer containing 100 mM KCl at 30 thour followed by successive elution with buffers containing 250 mM KCl (80 ml at 40 ml/hour), 600 mM KCl (60 ml at 30 ml/hour), and 1 M KCl (40 ml at 60 thour). The enzyme eluted at 400 mM KCl (Figure 1B) and the active fractions were pooled (fiaction III). Octyl-Sepharose Chromatography - To increase hydrophobic interactions, solid ammonium sulfate was added to fraction 111 (0.36 g/ml) over 30 min, and the suspension 49 was stirred for an additional 20 min. The suspension was then loaded onto an octyl- Sepharose column (0.6 x 2.5 cm) which was equilibrated with 20 mM potassium phosphate buffer at a flow rate of 1.4 ml/hour. The column was washed with 2.8 ml of equilibration buffer at 2.1 m1/hour and then eluted at the same flow rate with 2.5 ml of buffer containing 0.3% Triton X-100, 2.8 ml of buffer containing 1% Triton X-100, and 2.1 ml of buffer containing 2% Triton X-100. The enzyme eluted after application of the buffer containing 1% Triton X-100 (fraction IV). While this step results in only a 1.2 to 1.8-fold increase in specific activity, it is required both to de-salt and to concentrate the enzyme fraction before further purification. Cibacron Blue-Agarose Chromatography - Fraction IV was applied directly to a Cibacron Blue-agarose column (0.6 x 2.8 cm) equilibrated with 20 mM potassium phosphate buffer containing 9% glycerol and 0.015% Triton X-100. The column was washed with 1.6 ml of buffer containing 50 mM KCl at 1.2 ml/hour, followed by 2.4 ml each of buffers containing 100 mM, 350 mM, 1 M and 2 M KCl at 2.4 mllhour. Enzyme activity eluted between 400-550 mM KCl (fraction V, Figure 1C). Glycerol Gradient Sedimentation - Fraction V was layered onto two pre-formed 10- 30% glycerol gradients containing 50 mM potassium phosphate, pH 7 .6, 200 mM (NH4)2SO4, 0.015% Triton X-100, 2 mM dithiothreitol, 2 mM EDTA, 1 mM PMSF, 10 mM sodium metabisulfite and 2 ug/ml leupeptin, prepared in polyallomer tubes for use in a Beckman SW41 rotor. Centrifugation was at 37,000 rpm for 60 h at 3°C, after which 8 drop fractions were collected Active fiactions were pooled, an equal volume of 25 mM HEPES, pH 8.0, 2 mM EDTA, 80% glycerol and 0.015% Triton X-100 was added, and the enzyme (fraction VI) was stored at -20°C, -80°C or under liquid nitrogen. Gel Filtration - DNA polymerase 7 (fraction VII, 19 units) was chromatographed at a flow rate of 0.9 mllh on a Sephacryl S-200 column (0.85 x 47 cm) equilibrated with 50 mM potassium phosphate buffer containing 200 mM (NH4)2SO4 and 0.015% Triton X- 100. Fractions (0.14 ml) were collected and aliquots (2 pl) were assayed for DNA 50 Figure 1. Purification of the Dmsgphila melanogaster DNA polymerase ‘y. A. Phosphocellulose chromatography. The mitochondrial extract (fraction I) derived from 200 g of Drgsgphila embryos was chromatographed on phosphocellulose P- 11 as described under "Methods." The recovery of enzyme activity in the pooled fractions 26-34 was 56% (fraction H). B. Single-stranded DNA-cellulose chromatography. Fraction H was fractionated with ammonium sulfate (fraction IIb) and then dialyzed and chromatographed on DNA-cellulose as described under "Methods". The recovery of enzyme activity in the pooled fractions 23-27 was 54% (fraction III). C. Cibacron Blue-agarose chromatography. Fraction [II was de-salted and concentrated by chromatography on octyl-Sepharose (fraction IV), and then chromatographed on Cibacron Blue-agarose as described under "Methods". The recovery of enzyme activity in the pooled fractions 13-24 was 85% (fraction V). 51 0 Nb. . 39. .35.: 2 11-2... 2052,5328 » mama. / 0 J 20 2 I 01:23 >t>Fo< x .5“. FRACTION NUMBER 52 polymerase activity. The column was calibrated with blue dextran 2000, bovine liver catalase, rabbit muscle phosphorylase b, and yeast alcohol dehydrogenase. Gel Electrophoresis and Protein Transfers - Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate was performed according to Laemmli (18). The proteins were stained in the gel with silver (19). Alternatively, the proteins were transferred electrophoretically from the polyacrylamide gel to nitrocellulose paper using a Hoefer Transphor apparatus under the conditions suggested by the manufacturer, and then stained with iodine-starch (20). Analysis of DNA Polymerase Activity in Sim - The method for analysis of DNA polymerase activity in sjm was essentially that of Hiibscher (21). The enzyme sample in 65 mM Tris-HCl, pH 6.8, 2 mM EDTA, 125 mM 2-mercaptoethanol, 10% glycerol, 1% SDS and 10 ug heterogeneous protein mixture (22; a mitochondrial extract which was inactivated by incubation at 100°C for 4 h) was heated at 37°C for 3 min and electrophoresed in a 7.5% (w/v) polyacrylamide gel containing 0.1% SDS, 2 mM EDTA and activated calf thymus DNA at a final concentration of 125 ug/ml. Electrophoresis was at 20 volts for 3 hours at 25°C. SDS was removed from the gel by immersing it twice in 30 gel volumes of 50 mM Tris-HCl, pH 8.5 at 25°C for 15 min. The proteins were allowed to renature in 5 gel volumes of 50 mM Tris-HCl, pH 8.5, 5 mM MgC12, 200 mM KCl, 400 ug/ml BSA, 16% glycerol, 0.01 mM EDTA and 10 mM 2-mercaptoethanol at 4°C for 4 h. The gel was then incubated in 5 volumes of the same buffer containing 15 BM dATP, 15 LLM dGTP, 0.3 uM dCTP, 0.3 LLM dTTP, 50 uCi ct[32P] dTTP, and 50 uCi (1(32P] dCTP at 30°C for 20 h. Unincorporated deoxyribonucleoside triphosphates were washed out by incubation in 5% (w/v) trichloroacetic acid containing 1% (w/V) sodium pyrophosphate at 4°C for 60 h. The gel was dried and autoradiographed at -80°C for 14 h on Kodak XAR-S film using a Dupont Cronex Quanta III intensifying screen. 53 Protein Determinations - Protein was determined by the method of Bradford ( 16) or Schaffner and Weissmann (17) as indicated. Bovine serum albumin was the protein standard. BESLZLIS. Purification of Mia DNA polymerase 7 To determine the most abundant source of the mitochondrial DNA polymerase, a survey of six developmental stages of 12131553211113 was undertaken. DNA polymerase 7 was assayed after isolation and extraction of mitochondria as described under "Methods." Embryos yielded 16 to 180-fold greater activity than organisms harvested at other developmental stages (Table 1). Correspondingly, the specific activity of DNA polymerase 7 was increased 5 to 30-fold in embryos relative to more highly developed Mia. Employing embryos as the starting material, various methods of extraction of the activity from mitochondria were investigated including sonication, blending with glass beads, osmotic lysis and incubation in the presence of NaCl (0.05-1.0 M) and/or detergent (0.05-2% sodium cholate or Triton X— 100). Optimal and reproducible yields of DNA polymerase 7 were obtained upon incubation in buffer containing 0.3 M NaCl and 2.0% sodium cholate. Notably, we observed a ZOO-fold dependence on salt and detergent relative to extraction of the mitochondria in low ionic strength buffer alone, thus providing preliminary evidence that the activity isolated is mitochondrial. The mitochondrial DNA polymerase from D. melanogaster embryos was purified 2500-fold with a yield of 3%. A typical purification is summarized in Table 2. The procedure yielded 10 pg of near-homogeneous enzyme from 200 g of embryos. In contrast, we obtain 0.5 mg of DNA polymerase (I from the same starting material with Table 1 54 DNA polymerase 7 activity during development of D. W embryosa Developmental Mitochondrial Specific we Age protein AW hour trig/gm units/gm units/mg % trssue trssue protern Embryos 0-16 5.4 35.9 6.6 100.0 lst instar larvae 22-27 0.8 1.0 1.3 2.8 2nd instar larvae 51-56 1.1 0.4 0.4 1.1 3rd instar larvae 96-101 1.0 0.2 0.2 0.6 Pupae 139-147 2.1 2.3 1.1 6.4 Adult, 4 days old 2.6 2.0 0.8 5.6 aA mitochondrial extract (fraction 1') was prepared from organisms harvested at the indicated stages of development, and DNA polymerase 7 activity was assayed under standard conditions as described under "Methods". 55 Table 2 Purification of DNA polymerase 7 hour D. W embryos Specific Fraction Volume Protein thivity activity Yield m1 mg units units/mg % I. Mitochondrial extracta 254.0 813 8460 10.4 100 II. Phosphocellulose and 3.0 10.2 2200 216 26.1 ammonium sulfate III. DNA-cellulose 31.0 0.6 1 180 1,970 14.0 IV. Octyl-Sepharose 2.8 0.29 608 2,100 7.2 V. Cibacron blue-agarose 1.3 0.10 516 5,160 6.1 VI. Glycerol gradient 2.7 0.01 269 26,900 3.2 a Fraction I was prepared fiom 200 g of embryos. 56 approximately the same yield (2), indicating that the abundance of the mitochondrial enzyme is about 50-fold less than that of the major replicative enzyme in early embryos. Physical Properties Electrophoresis of the fraction VI enzyme in an SDS- polyacrylamide gel revealed two major polypeptides with Mrs of 125,000 and 35,000 (Figure 2). Densitometric scanning after transfer of the polypeptides to nitrocellulose paper and staining with iodine- starch, or after staining in gig; with silver indicated that the relative abundance of the two polypeptides by the two staining methods was 1:1 and 1:0.9, respectively. Glycerol gradient sedimentation of the near-homogeneous enzyme (fraction VI) in the presence of 0.2 M (NI-l4)2SO4 yielded a single peak of DNA polymerase activity with a sedimentation coefficient of 7.6 8 (Figure 3). Values of 7.5 to 7.7 S were obtained in 4 determinations on 3 preparations. The native molecular weight of the mitochondrial DNA polymerase was estimated by combining the sedimentation coefficient with the Stokes radius as determined by Sephacryl S-200 gel filtration in the presence of 0.2 M (NH4)2SO4 (24). The DNA polymerase 7 (fraction VI) had an observed Stokes radius of 51 A (two determinations, Figure 4). Combining the sedimentation and gel filtration results yields a calculated native molecular weight of approximately 160,000 daltons (Appendix A). Moreover, SDS- polyacrylamide gel electrophoresis of the peak fractions fiorn the gel filtration column again revealed the 125,000 and 35,000 dalton polypeptides (Figure 2, lane 3). Thus, the data indicate that the W DNA polymerase ‘y is most likely a heterodimer. 57 Figure 2. SDS-polyacrylamide gel electrophoresis of D. melanogaster DNA polymerase 7. Fraction VI, prior to (68 units, lane 1; 25 units, lane 2) and after (11 units, lane 3) chromatography on Sephacryl S-200, was denatured and electrophoresed in a 5-15% linear gradient SDS-polyacrylarrride slab gel. Protein was detected by iodine-starch (lane 1) or silver (lanes 2 and 3) staining as described under "Methods". Marker proteins electrophoresed in adjacent lanes and indicated by their molecular weights (x 103) were: E. coll RNA polymerase B subunit, 1;. col; B-galactosidase, rabbit muscle glycogen phosphorylase b, bovine serum albumin, rabbit muscle pyruvate kinase, E. cgfi alkaline phosphatase, E. col; RNA polymerase 0t subunit, and bovine carbonic anhydrase. 125 ..g —35— _ 58 —155 —ll6 —93 59 Figure 3. Glycerol gradient sedimentation of Drosgphila DNA polymerase 7. DNA polymerase 7 (fraction V) was sedimented in a 10-30% glycerol gradient as described under "Methods". The yield was 52%. Protein markers run in parallel gradients were: bovine liver catalase (CAT, 11.3 S), rabbit muscle L-lactate dehydrogenase (LDH, 7.3 S), and E. M DNA polymerase I (POL I, 5.5 S). 60 R3 I POL 7 ACTIVITY pmoI/ulxlol A m l O l 20 3O 4O FRACTION NUMBER I I I I j _ 15 {'3 IO CAT 3 ' LOH .. g, . > POL I m 0 0 IO 20 30 4O FRACTION NUMBER 50 61 Figure 4. Gel filtration of mm DNA polymerase 7. DNA polymerase 7 (fraction VI, 19 units) was chromatographed on Sephacryl S-200 as described under "Methods". The yield was 58%. Fractions containing the peak of activity were pooled as indicated by the bracket and analyzed by SDS-polyacrylamide gel electmphoresis (Figure 2, lane 3). The protein markers indicated were yeast alcohol dehydrogenase (ADH, 46 A), rabbit muscle phosphorylase B (PHOS b, 48 A), and bovine liver catalase (CAT, 52 A). Kd= (V e-Vo)/(V t-Vg-Vo). Ve, elution volume corresponding to the peak concentration of the solute; Vo, void volume of the column (elution volume of a substance which does not penetrate the solvent space interior to the gel grains); Vg, volume not accessible to the solvent, Vt, total volume of the gel bed (From: Siegel, L. M. and Monty, K. J. (1966) Biochim. Biophys Acta. 112: 346-362) (POL y ACTIVITY pmol/pl d) .5 N O 62 0.0 10 Q ADH L PHOS b u — 2‘, 05 CAT 0 1 40 so so 0.I Stokes' radius. A 0.15 6 3 DNA polymerization i_n_ Sim To determine which of the two polypeptides associated with the near-homogeneous enzyme is responsible for the DNA polymerase activity, the crude (fiaction I) and the highly purified (fraction VI) enzymes were subjected to an activity gel analysis. The procedure involves denaturation and electrophoresis of the enzyme in an SDS- polyacrylamide gel, followed by renaturation and enzyme assay in sin; (21). The data in Figure 5 demonstrate that the DNA polymerization function may be associated with the larger of the two polypeptides (125,000 daltons) in both enzyme fractions. On the other hand, no polymerization activity can be ascribed to the 35,000 dalton polypeptide. Furthermore, the molecular weight of the catalytic core polypeptide was identical in the mitochondrial extract (Figure 5, lane 2) and in the near-homogeneous enzyme (Figure 5, lane 3), indicating that the 125,000 dalton species most likely represents an unproteolyzed subunit of the mitochondrial enzyme present '31 m. Reaction Requirements Maximal activity on activated calf thymus DNA required the four deoxyribonucleoside triphosphates, Mg2+ and KC]. The KC] optimum was 200 mM (Figure 6), clearly distinguishing the mitochondrial enzyme from DNA polymerase a which has a KC] optimum of 50 mM (25). The optimal pH in 50 mM Tris.HCl buffer was 9.0: the reaction rates at pH 6.5 and 10.0 were 55% and 71% of optimal, respectively. A divalent metal cation was absolutely required for activity: Mg2+ at its optimum of 24 mM stimulated the polymerase 7-fold more than Mn2+ at its optimum of 0.5-2 mM. The apparent Km for dTTP was 0.97 i 0.1 M for both the crude (fiaction I) and the near- homogeneous enzyme (fiaction VI), indicating that native structure of the enzyme is most likely retained during purification (Figure 7). Fraction VI exhibited a linear reaction rate for 40 min at 30°C (data not shown). 64 Figure 5. DNA polymerization by crude and near-homogeneous 121259911112 DNA polymerase y i_n sign. DNA polymerase 7 fraction 1 (0.6 units, lane 2) and fraction V1 (0.6 units, lane 3) were denatured, electrophoresed in a 7.5% SDS- polyacrylamide gel, renatured and assayed for DNA synthesis i_n 5131 as described under "Methods". E. 9911 DNA polymerase I (109,000 daltons) and its Klenow fragment (76,000 daltons) were used as DNA polymerization standards, and as internal molecular weight markers (lane 1). Other protein markers for molecular weight were as indicated in the legend to Figure 2. 65 66 Figure 6. Dependence of DNA polymerase 7 activity on monovalent cation concentration. DNA polymerase y (fiaction VI) was assayed under standard conditions in the presence of the indicated concentrations of KC]. 67 — _ - l00 _ _ AU O 2 3.2.3 >t>fio< x mm 0 0 uOd [KCI], mM 68 Figure 7. Determination of Km for dTTP of the crude and near-homogeneous Drosgphila DNA Polymerase 7. DNA polymerase 7 was assayed under standard conditions except that the dTTP concentration was varied The data is presented in the form of a Lineweaver-Burk plot. Fraction I (0); Fraction VI (0). 69 I/V 70 To distinguish clearly the activity of the Mgph—ila melanogaster DNA polymerase y from those of D. melanogaster DNA polymerase a (2) and D. melanogaster DNA polymerase B (6), effects of the nucleotide analog dideoxy TI'P (dz'I'I‘P) and the sulflrydryl group blocking agent N-ethylmaleimide (NEM) were examined. The data in Figure 8 indicate that the DNA polymerase 7 activity is both strongly inhibited by d2'I'I'P (53% inhibition at 5 uM dzTTP/BO ttM dTTP), and by NEM (81% inhibition at 0.2 mM). On the other hand, while DNA polymerase or activity is inhibited by NEM, it is not inhibited by d2'ITP (Figure 8, ref. 25); DNA polymerase B activity is inhibited by dz'l'l'P but not by NEM (6). Thermal Stability and Role of DNA, dNTPs and ATP Drgsophila DNA polymerase yretains ~60% of its catalytic activity upon preincubation for 40 minutes at 30°C under standard assay conditions in the absence of DNA and dNTPs. In order to assess the potential stabilizing and/or stimulatory roles of various factors, we examined their effects under conditions of preincubation at 37°C prior to assay at 30°C. In the absence of DNA and dNTPs, DNA polymerase yexhibited 35% residual activity after 2 minutes, and only 0.2% activity after 20 minutes of preincubation at 37°C (Figure 9A). The addition of dNTPs to the preincubation mixture yielded only a small increase in enzyme stability. However, DNA provided a ZOO-fold increase in enzyme stability such that 49% of the catalytic activity was retained after 20 minutes of preincubation at 37 °C. Because the DNA polymerase likely binds the template-primer prior to nucleotide binding it is perhaps significant that dNTPs provide a 1.2- to 2.0-fold stabilization over DNA alone at times exceeding 20 minutes of preincubation. In evaluating further the effect of DNA on the thermal stability of DNA polymerase y, we found that at 40 minutes of preincubation the enzyme retained 100% activity at 30°C, 73% at 34°C and 19% at 37°C, but was completely inactivated by‘preincubation at 42°C (Figure 9B). 71 Figure 8. Inhibition of the highly purified Drgsgphila melanogaster DNA polymerase 7 activity by dideoxythymidine triphosphate and N—ethylmaleimide. Assays for inhibition by d2'I'I'P and NEM were performed under standard conditions with the following exceptions. The final KC] concentration was 200 mM for DNA polymerase 'y (0), and 20 mM for DNA polymerase 0t (0) and DNA polymerase I (l). A. The final dTTP concentration was 30 BM, while the d2'ITP concentration was varied as indicated. B. To determine inhibition by NEM, 2- mercaptoethanol was omitted from reaction mixtures, and the complete reaction mixtures including the enzyme, and containing the indicated concentrations of NEM, were pre-incubated for 30 min at 0°C prior to incubation for 20 min at 30°C. RELATIVE ACTIVITY °/o 72 0 0 2'0 I 410 60 o [dzTTP], ,tM 73 Figure 9. A: Thermal stability of DNA polymerase 7. DNA polymerase 7 (fraction VI, 0.12 unit) was assayed under standard conditions on DNase I- activated calf thymus DNA for 20 min at 30°C, except that a preincubation at 37°C was performed in assay buffer lacking DNA and dNTPs with the following additions: o-o, none; °--, + DNA; C1-C1, + dNTPs; I-I, + dNTPs, DNA; A - A, + ATP. B: DNA polymerase 7 was assayed as indicated above except that the DNase-I activated calf thymus DNA was included in the preincubation buffer and preincubation was performed at the following temperatures: o-o, 30°C; 0 - 0, 34°C; D-C], 37°C; I-I, 42°C. RELATIVE ACTIVITY, °/o RELATIVE ACTIVITY, °/o 74 37°C I00 (I) O / - / 60' ‘ 4o -\ 0 20 - e \ 0 ‘II— - ~----—-—-- - r 0 IO 20 30 40 so 60 TIME OF PREINCUBATION, MINUTES us 3 ‘ ' +DNA . I + r 0 I0 20 30 40 50 60 TIME OF PREINCUBATION, MINUTES 75 Although ATP is not a substrate for DNA polymerase y, we investigated its potential stabilizing and stimulatory effects in part because the ATP concentration in mitochondria is high. Further, it has been demonstrated that ATP activates E. Elli. DNA polymerase III holoenzyme via the formation of a reactive initiation complex (26), and that ATP both stimulates and stabilizes calf thymus DNA polymerase 0t (27). Our data indicate that ATP (1 mM) does not increase the thermal stability of DNA polymerase 7 (Figure 9A). Further ATP does not stimulate DNA polymerase activity under standard assay conditions on activated calf thymus DNA, singly-prim 0 X174 DNA or poly (dA)-oligo (dT) when it is included in the reaction mixture at levels ranging from 0.05-5 mM. Template-primer Specificity Purification of the mitochondrial DNA polymerase was based on its DNA synthetic activity on DNase I-treated (activated) calf thymus DNA, which contains nicks and short single-stranded DNA gaps. The activity of DNA polymerase yon ribo- and deoxyribo- homopolymers containing single-stranded DNA regions of average length of 200 nucleotides was examined (Table 3). Though the KC] optimum was shifted from 200 mM to 80 mM, the relative activity of the enzyme as compared to activated calf thymus DNA was 2.1 and 5.5-fold greater on poly (rA)-p(dT)1() and poly (dA)-p(dT)10, respectively. Further, on ¢X174 DNA (5386 nucleotides) primed with a unique lS-mer, synthesis was equivalent to that on activated DNA, demonstrating that the D. W DNA polymerase y utilizes efficiently template-primers with widely varying primer densities: the inter-primer distance among the substrates examined ranges from several nucleotides in the activated calf thymus DNA, to 5000 nucleotides in the singly-primed ¢X174 DNA. The accompanying shift in the KC] optimum upon incubation of DNA polymerase y with template-primer DNAs containing low primer to single-stranded DNA ratios might relate either to template structure (with regard to the accessibility of the single-stranded DNA regions to be c0pied), or to conditions chuired for efficient binding of the enzyme to the 76 Table 3 Template-primer specificity of DNA polymerase y from D. melanogaster embryosa I 1 _ . E . . : . I I: 200 mM 80 mM 20 mM KC1 KC] KC] Activated calf thymus DNA 35.5 2.1 0.2 Poly (rA) - p(dT)10 0.2 74.8 54.8 Poly (dA) - p(dT)10 1.3 195 56.8 Singly-primed ¢X174 DNA 0.5 42.7 16.6 Unprimed ¢x174 DNA X174 . 1 l r l l l l l 20 40 60 80 TIME, minutes 92 Enzyme activity was ~2-fold less at all time points on DNase I-activated DNA, but the affinity of the enzyme for the latter substrate is higher than for the former (Table 1). As a consequence, DNA polymerase y exhibits a relative substrate specificity (kcat/Km) of ~1.0 for these two template-primers of high primer density. DNA polymerase activity on activated DNA was approximately two-fold greater than that observed on the two natural single—stranded DNAs for which the enzyme exhibited a significantly higher affinity (lower Km values), resulting in an ~10-fold greater subsu'ate specificity. The data suggest that the template features of nucleotide composition (or DNA sequence specificity), single-strandedness and primer density influence catalysis by DNA polymerase 7. DNA polymerization on singly-primed 0X174 DNA by ypolymerase is approximately 50% relative to that on activated calf thymus DNA. This is dramatically higher than that observed with the replicative a polymerase which exhibits only a 3-8% relative efficiency (12, this study). The products of DNA synthesis on the singly-primed aXl74 DNA were examined by polyacrylamide gel elecuophoresis under denaturing conditions, and compared to those obtained upon incubation with E. 5311 DNA polymerase I which exhibits similar template usage properties. Under conditions of substrate excess (~25 primer termini per enzyme molecule), the extents of DNA synthesis for the two enzymes were similar at each time point: after incubation for 80 min, 12.2% of the available template was copied by DNA polymerase y, and 10.8% by DNA polymerase I (Figure 2 legend). At the same time, a remarkable difference in the mechanism of DNA synthesis by the two enzymes was evident. DNA strands greater than 700 nucleotides in length were synthesized by DNA polymerase 7 after only 2 min of incubation, when the extent of DNA synthesis was 0.3%, and the calculated average product size was 16 nt (Figure 2, lane 1). In contrast, nascent DNA strands larger than 600 nucleotides in length were not observed until 20 min of incubation with DNA polymerase I when the extent of DNA synthesis was nearly 10-fold greater, at 2.6% (Figure 2, lane 11). In the DNA polymerase yreaction, the abundance of 93 Table 1 Kinetic parameters of template-primer utilization by DNA Polymerase 1° Template-primer KC] optimum Km kcat mM uM nt inc. seC'1 enzyme molecule'1 Poly(dA)-p(dT) 120 30.3 i 6.8 6.75 :l: 0.81 Activated calf thymus DNA 200 12.5 :I: 3.0 3.08 :t 0.32 Multi-primed M13 DNA 110 0.57 2]: 0.17 2.41 :I: 0.21 Singly-primed aXl74 DNA 120 1.06 i 0.27 1.66 :I: 0.81 3P0] 7 (fraction VI) was assayed under standard conditions except that the template- primer concentration was varied, and the optimal KC] concentration for each template-primer was used. Incubation was for 5 min at 30°C. Other experimental details were as described under "Experimental Procedures." 94 Figure 2. Analysis of the products of DNA synthesis on singly-primed 9X174 DNA. The DNA products obtained upon incubation of Emsophila DNA polymerase 7 (0.63 unit) or E. 9911 DNA polymerase I (0.008 unit) with singly- primed 9X174 DNA (12 11M) were isolated, denatured and electrophoresed in a denaturing 6% polyacrylamide gel as described under "Methods." The time intervals and extents of synthesis for the DNA polymerase yreactions were: 2 min- -0.3%, 5 min--1%, 10 min--2%, 20 min--5%, 40 min--7%, 60 min--9% and 80 min-- 12% (lanes 1-7, respectively). Those for the DNA polymerase I reactions were: 2 min--0.3%, 5 min--0.7%, 10 min--1.4%, 20 min--3%, 40 min--6%, 60 min--8% and 80 min--11% (lanes 8-14, respectively). Lane 15 contains HpaII restriction fragments of M13Gori1 replicative form (RF) DNA (13): fi'agments larger than 900 nt were not fractionated under these conditions and migrated together in the first band; the sizes of the other fragments are 818, 652, 543, 472, 454, 381, 357, 272, 176, 156, 129 and 123 nt, respectively. The sizes of several major product classes are indicated on the left. Fragments smaller than ~75 nt migrated off the gel. No full length products (5386 nucleotides) were observed. 95 l234567 BSDHRBMI5 . o o C C . 3|0 - a...“ . 245 — O '65 - ,"w . ”0 - ' ills .. 90- p “I, 96 the smallest observed product (110 nt) relative to large products (~600 nt) was similar at the 2 and 20 min time points: ratios of 2.2 and 1.3, respectively. On the other hand, in the DNA polymerase I reaction the relative abundance of large products increased dramatically in the same time interval; the ratio of the 90 nt product to large products (~600 nt) was >78 at 2 min, and 3.7 at 20 min. The data suggest that the W mitochondrial DNA polymerase preferentially utilizes previously extended 3'-OH termini, perhaps by a mechanism involving incomplete dissociation after a polymerization cycle. Further, the presence of discrete product classes suggests that DNA polymerase y is sensitive to template secondary structure, a feature characteristic of DNA polymerase at from Q. melanogastg (14) and from vertebrate sources (15). DNA polymerase 7 exhibited a lower turnover number on the multi- and singly- primed single-stranded DNA templates as compared to the DNAs of high primer density under conditions of substrate excess (Table 1). We examined the possibility that impediments such as template secondary structure in the natural single-stranded DNAs might influence template-primer utilization under conditions of substrate limitation. The rate and extent of DNA synthesis were compared on the M13 DNA containing multiple primers at random sites and on the aXl74 DNA containing a single primer at a unique site. The data in Figure 3 indicate that the rate of DNA synthesis on both template-primers is nearly equal as was observed under conditions of substrate excess. Further, the extents of DNA synthesis of 85-100% after 80 min of incubation suggest that both template-primers can be copied completely by DNA polymerase 7. Previous work with the vaccinia virus (16) and bacteriophage T4 (17,18) DNA polymerases has indicated that several regions of the bacteriophage aXI74 and fd (homologous to M13) DNA genomes which may contain stable dyads cause dramatic decreases in the overall rate of DNA synthesis. Remarkably, catalysis by DNA polymerase 7 on 0X174 DNA is unaffected relative to polymerization on the multi-primed M13 DNA which has a 3-fold higher primer density and, because the primers are located at random sites, is a subsuate in which the potential inhibitory effects of 97 Figure 3. Time course of DNA synthesis in enzyme excess. DNA polymerase 7 fraction VI (1.26 units) was assayed on singly-primed aX174 DNA (0.4 11M) and multi-primed M13 DNA (0.4 11M). NUCLEOTIDE INCORPORATED, pmol 2.0 I6 I2 98 multi-primed M13 40 60 TIME, minutes singty-primed 3x174 80. 99 stable dyads are minimized (19). Because both template-primers contain extensive single- stranded DNA stretches and because DNA polymerase y has a high affinity for single- stranded DNAs of natural sequence (Table I), non-productive binding to regions without primer termini might occupy most of the enzyme molecules in either case. Thus, perhaps neither primer binding nor pausing during elongation is the rate limiting process, but rather dissociation of the enzyme from single-stranded DNA regions. The products of DNA synthesis on the two template-primers were analyzed by gel electrophoresis under native and denaturing conditions. Analysis of the reaction products on singly-primed aXl74 DNA under non-denaturing conditions revealed a complex array of products, indicating a variety of nascent DNA strand sizes (Figure 4A). In addition, accumulation of discrete products after 2, 5, and 10 minutes of incubation (lanes 1-3) indicated either the formation of complex structures or templates containing product strands of discrete lengths. Form II molecules representing complete replication were observed after only 5 min of incubation (lane 2) where they represented ~1% of the total products. After 40 min of incubation, the completed form II molecules predominated (lane 5), consistent with the extent of replication of 73% as determined by acid precipitation. Notwithstanding the formation of complex structures and/or discrete products on sin gly-primed 9X174 DNA resulting from the presence of stable dyads in the template, the rate of DNA synthesis was nearly identical to that observed on the multi-primed M13 (Figure 3). However, in the analysis of the products of DNA synthesis on the M13 template under both native (Figure 4A, lanes 8-13) and denaturing (Figure 4B, lanes 2-7) conditions, there was no indication of the formation of discrete products, but rather a continuous array of partially replicated molecules (Figtue 4A) and nascent DNA strand lengths (Figure 4B) as expected for a template primed at random positions. In this case, the input DNA contained circular and linear molecules in approximately equal amounts, such that the expected products upon native gel electrophoresis are form 11 and form III DNAs, respectively, as shown (Figure 4A). 100 Figure 4. Analysis of the products of DNA synthesis under conditions of template- primer limitation. A. The DNA products obtained upon incubation of DNA polymerase 7 fraction VI (1.26 units) with singly-primed 0X174 DNA (0.4 11M) and multi-primed M13 DNA (0.4 11M) were isolated and electrophoresed in a 0.8% agarose gel. The time intervals and extents of DNA synthesis on the singly-primed 0X174 DNA were: 2 min--14%, 5 min--26%, 10 min--39%, 20 min--56%, 40 min--73% and 80 min--86% (lanes 1-6, respectively). Those on the multi-primed M13 DNA were: 2 min--20%, 5 min--36%, 10 min--50%, 20 min--62%, 40 min- 83% and 80 min--99% (lanes 8-13, respectively). Lane 7 contains HpaII restriction fragments of M13Goril RF DNA whose sizes are 1925 and 1204 nt. 880 is single- stranded circular DNA; SS] is single-stranded linear DNA; HI and 1111 are duplex circular and linear DNA, respectively. B. The DNA products obtained in the reaction on the multi-primed M13 DNA were denatured and electrophoresed in a 0.8% alkaline agarose gel (lanes 2-7). Lane 1 contains restriction fragments of M13 RF DNA (ClaI: 3512 and 2895 nt) and M13Goril RF DNA (Hpall: 1925, 1204 and 829/819 nt; the fragments smaller than 800 nucleotides were not resolved.) 101 A I23456789I01||2I3 B l234567 102 Because the M13 template is multi-primed, genome-length product strands are not expected upon denaturation. The data in Figure 4B show that the average size of the nascent DNA strands increased from ~500 nt after 2 min of incubation (lane 1) to the calculated maximum average size of ~1800 nt at 80 min (lane 7). In comparison, we examined the nascent DNA strands synthesized by DNA polymerase 'y on the sin gly-primed aXl74 DNA by polyacrylamide gel elecuophoresis under denaturing conditions (Figure 5). As predicted from the native gel analysis, products of discrete lengths were obtained. Notably, of the discrete products observed, several could be directly correlated with the positions of the predicted stable dyads which presented major impediments to DNA synthesis by the vaccinia virus (16) and bacteriophage T4 (17) DNA polymerases: the 3'-ends of product strands of 2175, 3130, 3775 and 1800 nt in the DNA polymerase 7 analysis map to barrier A (nucleotide position 3974-3956 in the 0X174 DNA sequence), barrier B (3020-2997), barrier C (2373-2308) and barrier D (4357-4329), respectively. In addition, 14/20 of the discrete products could be correlated with computer- predicted helical regions in 0X174 DNA. The remainder could represent sites where primary sequence determinants are responsible for template-directed pausing. Such sites were found to inhibit DNA polymerase 11 from W (14) and from vertebrate sources (15). These studies also demonstrated that E. 5:911 DNA polymerase I is less sensitive to template secondary structure and relatively insensitive to primary sequence determinants. Taken together, the data in Figures 2 and 5 suggest that DNA polymerase 'y is sensitive to both primary sequence and secondary structure in the DNA template, accounting for the dramatic differences in the pausing patterns observed for DNA polymerase y as compared to DNA polymerase I. In this analysis under conditions of enzyme excess, full length species appeared after only 2 min of incubation, representing 0.2% of the total product strands (lane 2). At this time, 14% of the input DNA had been replicated and the calculated average product size was only 750 nt. Although the most abundant product was 1000 nucleotides, representing 103 Figure 5. Analysis of the products of DNA synthesis in the presence of limiting amounts of singly-primed ch74 DNA. The DNA products obtained upon incubation of DNA polymerase 7 fraction VI (1.26 units) with singly-primed 0X174 DNA (0.4 M) were isolated, denatured and electrophoresed in a denaturing 4% polyacrylamide gel as described under "Experimental Procedtu'es." The time intervals and extents of DNA synthesis were: 1 min--7%, 2 min--14%, 5 min--26%, 10 min--39%, 20 min--56%, 40 min-~73%, 60 rnin--77%, 80 min--86% (lanes 2-9, respectively). The restriction fragment markers were: Lane 1, M13 RF DNA (ClaI: 3512 and 2895 nt); lane 10, 0X174 RF DNA (XhoI: 5386); and lane 11, M13Goril RF DNA (HpaII: 1925, 1204, 829/818, 652, 543, 472 and 454 nt). The lengths in nt of major product classes are indicated on the left. No products smaller than ~400 nucleotides were detected. To correlate DNA product classes with helical regions in the DNA template, the positions of predicted stable dyads in 0X174 DNA were determined using the Beckman Microgenie program in an IBM XT computer. 5386 2I50 I850 I225 I025 840 820 650 480 430 23456 789 I0 0 105 19% of the total products, discrete products ranging from 430-5386 nt were observed. In the interval between 2 and 20 min of incubation (lanes 3-6) the increase in abundance of an 1850 nt intermediate was 7-fold, whereas that of the 5386 nt full length product increased 125-fold. After 40 min of incubation when the extent of replication was 73%, the predominant product was full length (5386 nt). As with the analysis in substrate excess, the data suggest that DNA polymerase y preferentially utilizes previously extended primer termini. Further, the enzyme has the capacity to replicate completely and efficiently a single-stranded DNA which has substantial secondary structure. Processivity of Nucleotide Polymerization The high rate of DNA synthesis on singly-primed aXl74 DNA relative to that on activated calf thymus DNA suggested that the mitochondrial DNA polymerase might replicate long stretches of single-stranded DNA without dissociation. However, the presence of discrete short products in the gel analyses of the reactions canied out both in substrate excess and in limitation indicate that DNA polymerase y is not a highly processive enzyme. To examine this issue, the processivity of DNA polymerase y in replicating sin gle-stranded DNA templates was measrned on multi-primed M13 DNA by a kinetic method (10), and on singly-primed 0X174 DNA by direct analysis of the reaction products in a denaturing polyacrylamide gel. Processivity, defined as the number of nucleotides polymerized in a single binding event, was measured kinetically on the multi-primed M13 DNA in large substrate excess (19). Poly (dA)-oligo (dT) was used as the competitive inhibitor (Ki=10 11M). The data in Table 2 indicate that DNA polymerase y is a moderately processive enzyme, polymerizing on average two-fold more nucleotides than E. can DNA polymerase I. The relative cycling I time is a measure of the time between enzyme binding cycles in the absence and in the presence of 106 Table 2 Kinetic assessment of the processivity of nucleotide polymerization by DNA polymerase 'y ‘1 Number of Relative . . Enzyme determinationsb cychng trme PYOCCSSWWY (nt) Poly 6 4.4:]:1.6 29.1 i 6.2 P011 2 11.9i2.1 l3.7:I:0.5 a A detailed account of the mathematical analysis is presented in: Bambara, R. A., Uyemura, D. and Choi, T. (1978) J. Biol. Chem. 253: 413-423. b Reactions were performed in uiplicate as described under "Methods." 107 DNA polymerization, and indicates the static versus kinetic affinity of the enzyme for the template-primer (10). The value for DNA polymerase y is ~3-fold less than for DNA polymerase 1, indicating that in the absence of DNA synthesis DNA polymerase yrecycles at a greater rate than DNA polymerase I. Mechanistically, this would allow the mitochondrial enzyme to complete more polymerization cycles per unit time on a per molecule basis than DNA polymerase 1. While the kinetic method yields a value representing the average number of nucleotides added per binding event, the direct measurement of processivity by denanuing polyacrylamide gel electrophoresis demonstrates the shortest DNA product actually synthesized. Gel electrophoresis of the DNA polymerase yreaction products on singly- primed 9 X174 DNA yielded a minimum processivity value of 25-40 nt (Figure 6), in agreement with the kinetic measurement. However, the appearance of a complex mixture of products >200 nucleotides in length at the very low extents of DNA synthesis indicated in the legend to Figure 6, suggests that a significant number of enzyme-substrate interactions actually result in highly processive synthesis catalyzed by DNA polymerase 7. Because of the lack of resolution of the large products in this analysis, we cannot make a definitive statement regarding their abundance. In subsequent experiments in which all of the products have been accurately quantitated, we find that products >200 nucleotides in length comprise 40% of the total DNA product strands and that the average product length determined by the gel analysis is ~150 nt (data not shown). Neither the fraction of longer products nor the average product length change over an 8-fold range of primer concentration. In contrast to the result with DNA polymerase 7, the processivity value , obtained in the gel analysis for DNA polymerase I is 15-30 nt; a significant fiaction of products longer than 100 nucleotides is not observed (data not shown). 108 Figure 6. Gel analysis of the processivity of DNA polymerase 7. DNA polymerase 7 fraction VI (0.03 unit) was assayed on the singly-primed eX174 DNA (50 M). The reaction products after 5, 10, 20 and 40 minutes of incubation (lanes 1-4, respectively) were isolated, denatured and electrophoresed in an 8% denaturing polyacrylamide gel as described under "Experimental Procedures." The extents of DNA synthesis were: 5 min--0.003%, 10 min-001%, 20 min--0.02% and 40 min-004%. The product sizes indicated were determined by comparison with a dideoxy sequencing ladder generated by incubation of the sin gly-primed 0 X174 DNA with DNA polymerase I Klenow fragment (20, lanes A,C,G,T). 109 ACGT |234 1 l 0 Fidelity of i_n, m DNA Synthesis The fidelity of DNA synthesis is a major factor in determining the overall mutation rate i_n 213.2. Because the accumulation of nucleotide substitutions in animal mitochondrial DNAs is 5- to 10-fold greater than in nuclear DNA genomes (21,22), an evaluation of the accuracy of nucleotide polymerization by DNA polymerase 'y is of particular significance. To examine the fidelity of DNA synthesis by DNA polymerase y, we have utilized the method of Weymouth and Loeb (23) to measure the reversion of the 6X174am3 mutation to wild type and pseudo-wild type as a result of nucleotide misincorporation druing DNA synthesis i_n_ .ViEQ- Reactions were performed with both unbiased and biased deoxynucleotide pools. Parallel experiments were carried out with 12. W DNA polymerase 7. As indicated in Table 3 the average number of nucleotides polymerized by DNA polymerase 7 on each primer terminus was ~7-fold greater than the distance from the primer terminus to the am3 mutation which lies 177 nucleotides downstream. The extent of DNA synthesis was unaffected by the various dNTP pool bias conditions. Transfection with DNA replicated by DNA polymerase 7 under conditions where all foru' dNTPs were present at equal concentrations (unbiased pool) yielded revertants at a level of 1.8 x 10'6 (Table 3). These values were consistently 2-fold higher than those determined for the uncopied DNA control. Notably, the reversion frequency of eX174 DNA synthesized by DNA polymerase 7 was nearly identical to that obtained in parallel with DNA polymerase at (data not shown)-the control experiments with DNA polymerase at were in complete agreement with our earlier determinations (12). Likewise, using a pool bias where dATP was in 100-fold excess over dTTP during in m DNA synthesis, the number of revertants was increased dramatically and by a similar value when polymerization was catalyzed by presenting DNA polymerase yor DNA polymerase or. In poo] bias experiments where the concentrations of other pairs of nucleotides were altered, no large increase in the number of revertants was observed with either DNA polymerase. Because it is possible that some dUTP is present during in m DNA synthesis as a result 111 Table 3 Reversion frequency of 0X174am3 DNA synthesized in mm by DNA polymerase 7 Reaction Nucleotides Phage titer‘ Reversion conditiona polymerizedb am3 Wild type frequencyd (x 10°6) (x 10'2) Unbiased pool 1126 13.3 0.24 1.8 x 10-6 A>T 1178 16.9 23.7 1.4 x 10-4 (78)° G>A 1398 21.4 0.45 2.1 x 10-6 (1.2) G>C 1427 20.5 0.39 1.9 x 10-6 (1.1) C>T 1183 23.5 1.9 8.1 x 10-6 (4.5) 2‘The experiments were performed as described under "Experimental Procedures." In the unbiased poo] determination the concentration of all four dNTPs was 40 M. In the biased pool determinations the concentrations of the indicated nucleotides were 1000 and 10 BM, respectively. IJ'I‘he average number of nucleotides added per primer terminus cThe phage titers listed are averages derived from 10 experiments. Variation in the data between experiments was <2-fold. d'Ihe reversion frequency of the uncopied control was 1.1 x 10'6. eNumbers in parentheses represent biased/unbiased values. 112 of spontaneous deamination of dCTP, the C>T pool bias may result in an underestimate of the reversion frequency for both enzymes (24). Nevertheless, in the ch74 am3 reversion assay, the accuracy of polymerization on single-stranded DNA in XIEQ by the near- homogeneous mitochondrial DNA polymerase is nearly identical to that of the replicative DNA polymerase fiom the same source - Mild melanogaster embryos. At the same time, the M13 DNA polymerase (I is one of the most accurate at polymerases which has been examined (12,25). Further, the fidelity of these eukaryotic DNA polymerases is remarkably similar to that of E. 9911 DNA polymerase III holoenzyme (12,26), even though the subunit structure, reaction requirements, kinetics and processivities of the three enzymes differ greatly. DISCUSSION Comparative studies have demonstrated that vertebrate mitochondrial DNA polymerases from a variety of tissues and cell types exhibit differential template-primer usage in m. Partially purified ‘y polymerases from mouse myeloma (27), human cells (28-30), rat brain (31), rat liver (29,32) and chick embryos (33) catalyze DNA synthesis on template-primers of high primer density with the general order of substrate preference being poly(rA)-oligo(dT) > poly(dA)-oligo(dT) > activawd DNA > denatured DNA (lower primer density). Further, enzyme preparations from human placenta (34) and calf liver (35) were shown to replicate single-stranded viral DNAs in mm. Otu' previous work with the near- homogeneous DNA polymerase y from 1211239911113 melamgasm embryos demonstrated that this invertebrate mitochondrial enzyme exhibits similar properties with regard to template-primer utilization (4). In this report, we show that the 12195521211113 enzyme exhibits Km values similar to highly purified mouse myeloma DNA polymerase y (36) for two template-primers: the Km values on activawd calf thymus DNA was 12.5 as compared to 40.6 W (as nt), and that on poly(dA).oligo(dT) was 0.15 as compared to 0.07 11M (as 113 primer termini), respectively. We have further investigated DNA polymerase/template- primer interactions ill glue to determine what factors may contribute to efficient enzyme function in m. In an examination of DNA synthesis on several template-primers, both natural and synthetic, we have identified template features which may affect catalysis by DNA polymerase x nucleotide composition, primer density and single-strandedness. 12mm DNA polymerase 7 replicates both predominantly sin gle-stranded and double-stranded DNAs. The rate of DNA synthesis is highest on the synthetic homopolymer poly(dA)-oligo(dT)10. Notably, WmtDNA genomes are 74-80% A + T (37). In this regard, it has been proposed that mtDNAs may have become A + T rich during long term evolution as a result of a functional requirement (or preference) of mtDNA and/or RNA polymerase for A + T rich DNA (38). Old data indicates that the D. W mtDN A polymerase is a versatile enzyme with regard to template-primer usage. In fact, the ability of DNA polymerase 'y to copy singly-primed DNA with a G + C content of 45% (9X174 DNA, 7) is 5- to 10-fold greater than that of DNA polymerase 01 (12, this study) relative to the rate at which these enzymes replicate predominantly double- stranded calf thymus DNA. In exhibiting a high relative efficiency in the replication of single-stranded DNA, DNA polymerase y suits the unique requirements for DNA polymerization in mitochondria. Because of the highly asymmetric mode of mtDNA replication in m leading DNA strand synthesis occurs on a duplex template while lagging DNA strand synthesis, which occurs on the displaced parental DNA strand, involves replication of a predominantly (or entirely) sin gle-stranded DNA template. We have shown that DNA polymerase 7 is sensitive to template secondary structure, which might imply a requirement for other protein factors to facilitate lagging DNA strand synthesis in 11549- At the same time, the enzyme's ability to replicate completely the BX 174 DNA genome under conditions of enzyme excess indicates that a high local concentration of DNA polymerase 7 would in itself be sufficient to replicate completely mtDNA. Ftu'thermore, the formation of stable secondary structures in the 114 template strand for lagging DNA strand synthesis jg m is perhaps minimal, as a result of the high A + T content of the nritochondrial genome. DNA polymerase y as purified from 12131552lean embryos does not copy very long stretches of natural sin gle-stranded DNA without the formation of intermediates: kinetic and direct analysis of enzyme processivity indicates that DNA polymerase y incorporates ~30 nucleotides before pausing and/or dissociating from the template. As determined by the kinetic method, the average processivity of DNA polymerase y is similar to that of the replicative DNA polymerase (I (39) and E. 9911 DNA polymerase I (10) which incorporate 10-20 nucleotides per binding event . On the other hand, although it is not a highly processive enzyme as is E. 95211 polymerase III holoenzyme which can polymerize several thousand nucleotides on nattn'al DNA without dissociation (19,40), the direct product analysis demonstrates that DNA polymerase v has the capacity for preferential extension of previously utilized primer termini. This is evidenced both by the appearance of a substantial fiaction of replication products which are much longer than would be expected if all primer ends were extended equally, and by the abundance of products which are multiples of the minimal unit length product under conditions of large template-primer excess. Taken together, the results suggest a quasi-processive DNA synthetic mechanism for 1212591211113 7 polymmse- Several possible mechanisms for quasi-processive synthesis are plausible. First, the near-homogeneous enzyme may consist of two distinct classes of enzyme, processive and less processive. We have proposed that the Mk 7 polymerase is a heterodimer which comprises a 125,000 dalton catalytic subunit and a 35,000 dalton subunit of unknown function (4). The processivity of the heterodimer may be greater than that of the catalytic subunit. Further, it is possible that the catalytic subunit may exist in several subunit stoichiometries resulting in, for example, a processive holoenzyme, a quasi- processive subassembly and a core enzyme of low processivity. This situation has been observed in analyses of E. 99E DNA polymerase III (19). Alternatively, two enzyme 115 classes may be generated by a specific chemical modification or from the presence of a limiting "template association" factor in the enzyme preparation which interacts with the enzyme during polymerization. Second, DNA polymerase ymay be capable of kinetic pausing without dissociation, or may release the primer terminus transiently but remain bound to the template strand. Third, two classes of template molecules may be distinguishable by DNA polymerase ‘y--those with and without structural constraints to replication. In this regard, a highly purified mitochondrial DNA polymerase fiom mouse myeloma (27) and the near-homogeneous enzyme from chick embryos (41,42) were shown to be highly processive on the ribohomopolymer poly(rA).oligo(dT). Even though this synthetic polyribonucleotide bears no resemblance to the in 2'19 template, it may allow the evaluation of the intrinsic processivity of the enzyme in the absence of DNA context effects. The high rate of evolution of animal mtDNAs and in particular, the high frequency of A(—)T transversions in mm mtDNA, have led to the suggestion that there is continuous selection for A + T nucleotides at all sites in mtDNA where it is compatible with function (3 8). This might occur either by an increased mutation rate resulting from infidelity during replication, or by an increased frequency of fixation of mutations, or both. We have demonstrated that the overall fidelity of replication catalyzed by DNA polymerase y is nearly identical to that of the replicative DNA polymerase at from Mia (12). Specifically, 121939911113 DNA polymerase 7 does not misincorporate to yield AHT transversions at a higher rate than DNA polymerase at. However, we have examined base substitution only at a single site-the am3 locus in aXl74 DNA, and it is possible that the m DNA polymerase 7 would be less accurate in another assay system. For examme, E. eon DNA polymerase III holoenzyme (219 exhibits an accuracy similar to that of the W enzymes (12) in unbiased and A > T biased pools in the am3 reversion assay, while its accuracy is several-fold higher in an A>T biased pool in the aXl74 am16 reversion assay (43). 116 The mitochondrial DNA polymerase from W embryos exhibits a 10-fold higher fidelity than the human HeLa cell DNA polymerase y as measured in the same assay system (44). Likewise, direct measurement of nucleotide misincorporation into several synthetic template-primers under a variety of conditions indicated that partially purified DNA polymerase y from human placenta (45) and human fibroblasts and HeLa cells (46) were relatively inaccurate and exhibited error rates 3- to 20-fold higher than the homologous y polymerase. In contrast, in a forward mutational assay capable of detecting a spectrum of base substitution and frameshift mutations, the near-homogeneous chick embryo and a partially purified calf liver DNA polymerase y are highly accurate (47,48). Clearly, the issue of the accuracy of the mitochondrial DNA polymerase and its role in the evolution of animal mtDNA warrant further study. Acknowledgement--M. C. Conway prepared the DNase I digested M13 DNA and also assisted with the Km determinations presented in Table I. OKLA-Dub.) 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. REE ERENQES Clayton, D. A. (1982) Cell 28: 693-705. Wolstenholme, D. R., Goddard, J. M., and Fauron, C. M.-R. (1979) In: Extrachromosomal DNA (Cummings, D., Dawid, I. B., Borst, P., Weissman, S. and Fox, C. F., Eds), pp. 409-425, Academic Press, Inc., New York. Goddard, J. M.,and Wolstenholme, D. R. (1980) Nucl. Acids Res. 8: 741-757. Wemette, C. M. and Kaguni, L. S. (1986) J. Biol. Chem. 261: 14764-14770. Fansler, B. S. and Loeb, L. A. (1974) Methods Enzymol. 29: 53-70. Cunningham, R. P., DasGupta, C., Shibata, T.,and Radding, C. M. (1980) Cell 20: 223-235. Sanger, F., Coulson, A. R., Friedmann, T., Air, G. N., Barrel], B. G., Brown, N. L., Fiddes, J. C., Hutchinson, C. A. HI, Slocombe, P. M. and Smith, M. (1978) J. Mol. Biol. 125: 225-246. Matson, S. W., Fay, P. J. and Bambara, R. A. (1980) Biochemistry 19: 2089- 2096. Kaguni, L. S., Rossignol, J.-M., Conaway, R. C.,and Lehman, I. R. (1983) Proc Natl. Acad. Sci. U.S.A. 80: 2221-2225. Bambara, R. A., Uyemura, D. and Choi, T. (1978) J. Biol. Chem. 253: 413-423. Kunkel, T. A.and Loeb, L. A. (1979) J. Biol. Chem. 254: 5718-5725. Kaguni, L. S., DiFrancesco, R. A.and Lehman, I. R. (1984) J. Biol. Chem. 259: 9314-9319. Kaguni, J. and Ray, D. S. (1980) Gene 11: 207-218. Kaguni, L. S. and Clayton, D. A. (1982) Proc. Natl. Acad. Sci. USA. 79: 983- 987. Weaver, D. T. and DePamphilis, M. L. (1982) J. Biol. Chem. 257: 2075-2086. Challberg, M. D. and Englund, P. T. (1979) J. Biol. Chem. 254: 7820-7826. Roth, A. C., Nossal, N. G. and Englund, P. T. (1982) J. Biol. Chem. 257: 1267-1273. Huang, C. _-C. and Hearst, J. E. (1980) Anal. Biochem. 103: 127-139. Fay, P. J., Johanson, K. O., McHenry, C. S. and Bambara, R. A. (1981) J. Biol. Chem. 256: 976-983. 117 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 118 Sanger, F., Nicklen, S. and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U .S .A. 74: 5463-5467. Dawid, I. B. (1972) Devel. Biol. 29: 139-151. Brown, W. M., George, M., Ir.and Wilson, A. C. (1979) Proc. Natl. Acad. Sci. U.S.A. 76: 1967-1971. Weymouth, L. A. and Loeb, L. A. (1978) Proc. Natl. Acad. Sci. USA. 75: 1924-1928. Baas, P. D., van Teeffelen, H. A. A. M., Teertstra, W. R., Jansz, H. S., Veeneman, G. H., van der Mare], G. A. and van Boom, J. H. (1980) F EBS Lett. 110: 15-20. Reyland, M. E. and Loeb, L. A. (1987) J. Biol. Chem. 262: 10824-10830. Fersht, A. R. (1979) Proc. Natl. Acad. Sci. U.S.A. 76: 4946-4950. Matsukage, A., Bohn, E. W. and Wilson, S. H. (1975) Biochemistry 14: 1006- 1020. Knopf, K. W., Yamada, M. and Weissbach, A. (1976) Biochemistry 15: 4540- 4548. Bolden, A., Pedrali-Nay, G. and Weissbach, A. (1977) J. Biol. Chem. 252: 3351-3356. Robert-Guroff, M., Schrecker, A. W., Brinkrnan, B. J. and Gallo, R. c. (1977) Biochemistry 16: 2866-2873. Hiibscher, U., Kuenzle, C. C. and Spadari, S. (1977) Eur. J. Biochem. 81: 249- 25 8. Tanaka, S. and Koike, K. (1977) Biochim. Biophys. Acta. 479: 290-299. Bertazzoni, U., Scovassi, A. I. and Brun, G. M. (1977) Eur. J. Biochem. 81: 237-248. Kollek, R. and Goulian, M. (1981) Proc. Natl. Acad. Sci. USA. 78: 6206- 6210. Robertson, A. T., Bates, R. C. and Stout, E. R. (1983) Biochem. Biophys. Res. Comm. 117: 580-586. Matsukage, A., Yamaguchi, M., Tanabe, K., Taguchi, Y. N., Nichizawa, M. and Takahashi, T. (1983) In: New Approaches in Eukaryotic DNA Replication, (deRecondo, A. M. Ed). PP. 81-106, Plenum Press, New York. Fauron, C. M.-R., and Wolstenholme, D. R. (1976) Proc. Natl. Acad. Sci. U.S.A. 73: 3623-3627. Wolstenholme, D. R. and Clary, D. O. (1985) Genetics 109: 725-744. ' 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 119 Villani, G., Fay, P. J., Bambara, R. A. and Lehman, I. R. (1981) J. Biol. Chem. 256: 8202-8207. Burgers, P. M. J. and Kornberg, A. ( 1982) J. Biol. Chem. 257 : 11474-11478. Yamaguchi, M., Matsukage, A., and Takahashi, T. (1980a) Nature 285: 45-47. Yamaguchi, M., Matsukage, A., and Takahashi, T. (1980b) J. Biol. Chem. 255: 7002-7009. Fersht, A. R.and Knill-Jones, J. W. (1983) J. Mol. Biol. 165: 633-654 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. (1985) J. Biol. Chem. 260: 12866-12874. Kunkel, T. A. and Alexander, P. S. (1986) J. Biol. Chem. 254: 5718-5725. Chapter IV Processivity of the mitochondrial DNA polymerase from 1212mm melanogaster cmbryosz template-primer. reaction condition and DNA binding protein effects 120 AM These studies were initiated to identify factors that affect template-primer usage and processivity of the mitochondrial DNA polymerase fi'om Drasopmla embryos. The product distribution generated by DNA polymerase y on the synthetic homopolymer poly(dA)- oligo(dT) is not affected by the reaction pH in the range of 6 - 9 or the concentration of divalent cations examined, although these factors do affect the reaction rate. The DNA product size classes produced on the synthetic homopolymer are nearly identical to those generated on natural DNA except that discrete products are observed on the latter template- primer. These may reflect the presence of DNA template secondary structure or intrinsic enzyme pausing or a combination of both effects. Inclusion of a heterologous single- stranded DNA binding protein increases both the rate of DNA synthesis and the processivity of the enzyme on natrual DNA 2-3 fold but does not result in a completely processive enzyme. Variations in template-primer or enzyme concentration do not affect processivity. Further, there does not appear to be any contaminating accessory factor in the enzyme preparation that alters the processivity of the mitochondrial DNA polymerase in any manner. Thus, the unusual template-primer usage behavior under reaction conditions that are optimal for DNA synthetic rate appears to be an intrinsic function of the enzyme in its current highly purified state. However, the Dmmhjla enzyme is processive for several thousand nucleotides under low ionic strength conditions which are suboptimal for DNA synthetic rate. 121 W DNA synthesis is a complex enzymatic process requiring that the DNA polymerase bind the primer terminus, incorporate the complementary nucleotide and dissociate from the template-primer. Several factors may influence this catalytic cycle and the average number of nucleotides incorporated before dissociation of the enzyme from the primer terminus-the processivity of the DNA polymerase. These factors may include temperature, ionic su'ength and the composition of the particular DNA template-primer, and are known to influence the processivity of E. 9911 DNA polymerase I (1). Recently, changes in reaction conditions such as pH (2) and magnesium ion concentration (3) were shown to increase the processivity of calf thymus DNA polymerase (1. Reduction of both reaction pH and magnesium ion concentration resulted in large increases in the processivity of calf thymus DNA polymerase 01 and DNA polymerase 8 (4). Accessory protein components may also play an important role in the regulation of DNA polymerase and template-primer interactions. The E. 2211 single-stranded DNA binding protein allows increased processivity of E. 9911 DNA polymerase III (5), Dggsophila DNA polymerase (I (3) and its isolated catalytically active large subunit (6). Moreover, particular enzyme subassemblies may exhibit altered processivity values as demonstrated in the case of E. £911 DNA polymerase 111 (5,7). Currently, there is no data that addresses the influence of the nature of the template- primer, reaction conditions or accessory protein factors on the processivity and template- primer usage of the mitochondrial DNA polymerases. Much information relating to the mechanism of M113 mitochondrial DNA synthesis has been accumulated by examination of the physical state of the mitochondrial DNA dming the replication process . (8). This has allowed the development of models for mitochondrial DNA replication which may describe the events occurring in 2119. However, further insight into the mechanism of mitochondrial DNA replication will require analysis of the highly purified enzymatic 122 123 components necessary for the initiation and elongation reactions of rrritochondrial DNA synthesis, and reconstitution of these reactions in m. Our initial studies were directed at the purification and partial characterization of the mitochondrial DNA polymerase fiom 1213842133114 embryos (9,10). Next, we examined template-primer usage, and the kinetics, fidelity and processivity of nucleotide polymerization by the near-homogeneous DNA polymerase 70]). These studies revealed that the W3 enzyme possesses structlual and catalytic features that distinguish it from other well studied mitochondrial DNA polymerases. Notably, the Ma enzyme does not operate by a completely processive mechanism under conditions optimal for DNA synthetic rate but does have a limited ability to utilize previously extended primer-termini. Here we present a more detailed examination of the template-primer usage properties of DNA polymerase y and the factors that may affect its behavior in m. These findings are evaluated in relation to the template-primer usage properties cf other eukaryotic DNA polymerases and to the requirements for mitochondrial DNA replication in _vi_vg. W Materials Nucleotides and Nucleic Acids - Unlabeled deoxynucleoside di- and triphosphates were purchased from P-L Biochemicals; a[32P]dTTP,0t[32P]dATP and at[32P]ATP were from New England Nuclear. Poly (dA)5ooo and oligo (dT)15 were purchased from P-L Biochemicals and annealed by incubation at 37°C for 30 minutes in a nucleotide molar ratio of 100: 1 , respectively, such that the average sin gle-stranded DNA region between primers was 1500 nucleotides. E. c911 RNA polymerase (provided by J.M. Kaguni of this department) was used to synthesize poly(rA) on poly(dT)sooo (Midland Certified Reagent Co.). The poly(rA)sooo was purified by ethanol precipitation and was annealed with p(dT)15 in the same manner as described above. A recombinant M13 viral DNA (10,650 124 nucleotides) was prepared by standard methods and annealed to a homologous synthetic oligonucleotide (15 nucleotides-position 174-188). Enzymes - m DNA polymerase 7 fraction IV (2,100 u/mg) and fraction VI (27,000 u/mg) were prepared as described (9). Momma DNA polymerase at fraction VI was further chromatographed by FPLC Superose 12 gel filtration to yield fraction VII. Methods DNA polymerase y Assay - The standard reaction mixtures (0.05 ml) contained 50 mM Tris-HCl (pH 8.5), 4 mM MgC12, 20 mM dithiothreitol, 120 mM KC], 50 11M (as nt)poly (dA)-oligo (dT), 400 ug/ml bovine serum albumin, 30 11M a[32P]T'I'P (1000 cpm/pmol) and DNA polymerase 7 fraction VI. Incubation was at 30°C for 10 minutes. Specific modifications are described in the figure legends. For the time course analyses, the reaction volume was increased to 0.175 ml and incubation was for 5, 15 and 30 minutes (or as indicated) at which time samples were withdrawn from the reaction mixture and placed on ice. A portion of each sample was precipitated with 10% trichloracetic acid/0.1 M sodium pyrophosphate to allow determination of the amount of nucleotide incorporated into acid insoluble product during the reaction. One unit of activity is that amount that catalyzes the incorporation of 1 nmol of deoxyribonucleoside triphosphate into acid-insoluble material in 60 minutes at 30°C. The remainder of each sample was prepared for analysis of the synthetic DNA products. Analysis of Products of DNA Synthesis by Denatruing Gel Electrophoresis - Products to be analyzed by denaturing gel electrophoresis were made 1% in sodium dodecyl sulfate and 10 mM in EDTA, heated for 10 min at 65°C and precipitated with ethanol in the presence of 1 ug sonicated calf thymus DNA as carrier. The ethanol precipitates were resuspended in 80% formamide, 1 mM EDTA, 0.1% xylene cyanol, and 0.1% bromphenol blue. Aliquots containing approximately equal amounts of radioactivity (500-1000 cpm) were denatured for 5 min at 100°C and electrophoresed in a 6% 125 polyacrylamide slab gel (13 x 30 x 0.07 cm) containing 7M urea in 90 mM Tris-borate (pH 8.3) and 25 mM EDTA. Alternatively, DNA samples were electrophoresed in a 1.5% alkaline agarose gel containing 35 mM NaOH and 2 mM EDTA. The gel was then washed in distilled water for 20 minutes, dried under vacuum, and exposed for ~24-48 hours at - 80°C to Kodak XAR-5 x-ray film using a DuPont Quanta III intensifying screen. Quantitation of the product analysis data was performed by densitometric scanning of the autoradiographs using a Hoefer model GS 300 densitometer. The area under the peaks was determined either by computer integration analysis or by the cut and weigh method which produced identical results. The calculated area (or weight) was normalized to the nucleotide level to correct for the uniform labeling of the DNA products. BESIHJIB. DNA Synthesis on the Synthetic Homopolymers Poly (dA)-oligo (dT) and Poly (rA)-oligo (dT) Our earlier results indicated that DNA polymerase 'y efficiently utilized a variety of synthetic and natural DNA template-primers (9,11). The general order of template usage under saturating conditions was poly (dA)-oligo (dT) > activated DNA > multi-primed DNA ~ singly-primed DNA. There was no activity of the mitochondrial DNA polymerase on unprimed or denatured DNAs. We found that to make valid comparisons, the reaction conditions for use of a particular template-primer must be optimized with regard to salt (11) and divalent cation concentration. The results indicated that 45 mM Mng was optimal for synthesis on activated DNA, multi-primed and singly-primed natural DNAs (not shown) and poly (dA)- oligo ((11) (Figure 1A). In contrast to results with calf thymus DNA polymerase 5 (2), the optimal magnesium concentration did not vary with reaction pH (data not shown). 126 Figure 1. Synthesis on synthetic homopolymers and effect of divalent cation. Standard assay conditions were used with 50 11M (as nt)of either poly (dA)-oligo (dT) (0) or poly (rA)-oligo ((11) (o); the indicated concenuation of MgClz (panel A) or MnC12 (panel B) and 0.06 unit of DNA polymerase yfraction VI. POL y ACTIVITY, u/ml 127 128 Maximal synthesis on poly (rA)-oligo (dT) required MnC12 (Figure 1B). At 2 mM MgC12 the activity on poly (rA)-oligo (dT) was 80% of that on poly (dA)-oligo (dT), but at 5 mM M gC12 this activity was reduced 2.5-fold relative to the other template-primer (Figure 1A), a finding which was observed previously (9). The activity on poly (rA)-oligo (dT) was simulated 2-fold by the use of MnC12 (0.2 mM) rather than MgClz (2 mM; Figure 1B) while the activity on poly (dA)-oligo ((11) was decreased 2.4-fold at 1 mM MnClz relative to its optimum with MgClg. The activity on poly (rA)-oligo (dT) with 0.2 mM MnC12 is stimulated 13-fold over that obtained with poly (dA)-oligo (dT). Processivity on a Synthetic Template-primer and Effect of Reaction pH Previous examination of the products of DNA synthesis on natural DNA suggested a quasi-processive mechanism of DNA replication '3; yin. Kinetic studies suggested a low processivity but it was clear from gel analysis experiments that the enzyme generated long DNA products by continued extension of previously utilized primer termini (11). Subsequent experiments were directed at analysis of factors that may influence the processivity of the enzyme i_n m. We examined the processivity of the mitochondrial enzyme on the synthetic homOpolymer poly (dA)-oli go (dT) in order to compare the results to those obtained on natural DNAs of complex nucleotide sequence and potential secondary structtue. We had previously shown that the mitochondrial polymerase is sensitive to template-primer secondary structure, a phenomenon that complicates the direct analysis of processivity due to the generation of numerous discrete products. The use of the synthetic homopolymer resulted in a homogeneous product distribution. Using a large excess of poly (dA)-oligo (dT) (3000 primer ends per enzyme molecule) the reaction rates were linear but varied with pH (Figure 2A). The activity at pH 6 was only 40% of that at the optimum at pH 8, whereas the activity at pH 7 was 90%, and that at pH 9 was 70% of the optimal value. 129 Figure 2. Effect of reaction pH on the rate of synthesis and processivity of DNA polymerase 'y on poly (dA)-oli go (dT). The rate of DNA synthesis was determined as described under "Methods" except that 50 mM Tris-HCl was included at the indicated pH with 0.028 unit DNA polymerase 7 fraction VI in a total volume of 0.175 ml (panel A). The weighted average size (in nucleotides) of the products of processive DNA synthesis at each pH was determined as described under "Methods" (panel B). The reaction pH values were 6 (D), 7 (o), 8 (0) and 9 (I). 130 6 EXTENT OF SYNTHESIS, m. m m m w 82.8.2... £52m... 528E noise .mamzezfi do ezmexm , minutes TIME °/. XIo2 131 Alkaline gel analysis of the products of DNA synthesis revealed a distribution of product sizes ranging from 1 to ~1500 nucleotides (Figure 3). No discrete products were observed. Under each reaction condition the most abundant products were <120 nucleotides in length, constituting 70-80% of the total products. Products from 120-500 nucleotides in length comprised 15-25% of the total population, while less than 3% were in the 500-1500 nucleotide range. The data demonstrate that pH has little effect on the processivity of the mitochondrial DNA polymerase (Figure 2B). At each pH, the calculated average processivity varied less than 2-fold while the extent of DNA synthesis varied nearly lO—fold. Further, the variation in processivity at the lowest extents of synthesis was less than 30% (Figure 3, lanes 1, 4, 7, 9). In contrast, lowering the reaction pH from 7.5 to 7.0 caused a 5-fold increase in the processivity of fetal calf thymus DNA polymerase 8 (2). Reduction of both pH and MgClz' concenuation increased the processivity of calf thymus DNA polymerase at from ~20 to >300 nucleotides and had a similar but less dramatic effect on calf thymus DNA polymerase 8 (4). Processivity on Natural DNA and the Effect of KC] Concentration Our previous studies have shown that the mitochondrial DNA polymerase is stimulated by salt (9), and that the optimal monovalent ion concentration varies somewhat depending on the nature of the template-primer (11). To further examine the effect of reaction conditions upon the product distribution generated by DNA polymerase 7, DNA synthesis reactions were canied out in various levels of KC]. In the absence of KC] the rate of DNA synthesis was 15% of that obtained at the KC] concentration which is optimal for DNA synthesis on predominantly single-stranded DNAs of nattual DNA sequence (120 mM KCl). Use of 60 or 180 mM KC1 yielded a nucleotide polymerization rate that was 60% of the optimum, while that obtained at 210 mM KCI was 30% of the optimum value. 132 Figure 3. Analysis of the DNA products synthesized by DNA polymerase yon poly (dA)-oligo (dT) and the effect of reaction pH. DNA products synthesized at the indicated pH by DNA polymerase 7 were analyzed on a 1.5% alkaline agarose gel as described under "Methods." The extents of DNA synthesis (% nucleotide incorporated relative to total input nucleotide in pmol) obtained for each reaction after 5, 15 or 30 minutes of incubation were: pH 6 (lanes 1-2; 5 min not shown)-- 0.003, 0.01 and 0.03; pH 7 (lanes 3-5)--0.01, 0.04 and 0.07; pH 8 (lanes 6-8)-- 0.01, 0.04 and 0.07; pH 9 (lanes 9-11)--0.01, 0.03 and 0.05. The DNA size markers used were a[32P]dCMP end-labeled restriction fragments of 9 X174 RF DNA (Xho I: 5386 nt); M13 RF DNA (Cla I: 3512 and 2895 nt) and M13 Gori I RF DNA (Hpa II: 1925, 1207, 829/818, 652, 543, 472/454, 381, 357, 272, 176, 156, 129/123, 60, 60 and .18 nt. I23456789|Ol| 1 3 4 The effect of ionic strength on the processivity of DNA polymerase 7 was dramatic (Figure 4). At 210 mM KC1 where the rate of synthesis was 30% of that obtained at the optimal KC] concentration, no products longer than 300 nucleotides were generated (lanes ‘9-10). The result was similar when the KC] concenuaan was 180 mM (lanes 7 and 8). As the salt concenuation was decreased to 120 mM where the DNA synthetic rate was optimal (lanes 5-6), longer products were generated but only ~10% were greater than 300 nucleotides in length. At 60 mM KCI (lanes 3 and 4) the full length product of 10,650 nucleotides was generated (5% of the total product), as well as several intermediates with a major species of ~2500 nucleotides accounting for 43% of the total products. The presence of these intermediates may reflect the effect of template secondary structure. Finally, in the absence of KC] full length products were abundant (47%) and the only other product apparent was the intermediate of ~2500 nucleotides (lanes 1 and 2) which comprised the remaining 53% of the product DNA strands. The data indicate that there was less than a 2-fold variation in average product size when KC] was present at 120 mM or higher. However, lowering the concentration to 60 mM resulted in an 18-fold increase in average product size at nearly identical extents of DNA synthesis. Eliminating salt from the reaction caused a 40-fold average increase in product size and allowed generation of a substantial proportion of full length products even though the rate of DNA synthesis was 7 -fold below the optimum. . Thus, under conditions optimal for DNA synthesis DNA polymerase 7 is processive for ~150 nucleotides. However, this ionic environment appears to promote the premature dissociation of the enzyme from the template. Under low ionic strength conditions the enzyme is processive for thousands of nucleotides and in fact, can synthesize a complete copy of the M13 DNA template without dissociation albeit at a relatively low rate. 135 Figure 4. Effect of KC] on the DNA products synthesized by DNA polymerase y The reaction mixtures (0.175 ml) contained 50 BM singly-primed M13 DNA, 30 trM each dATP, dGTP, dcra and ot[321>]d'm> (3000 cpm/pmol) and enzyme (0.06 unit). The KC] concentrations, time of incubation and extents of DNA synthesis were: No KC] (lane 1--90 min, 0.035%; 2--180 min, 0.05%), 60 mM KC] (lane 3--25 min, 0.02%; 4-50 min, 0.04%), 120 mM KC1 (lane 5--15 min, 0.025%; 6--30 min, 0.043%), 180 mM KC1 (lane 7--20 min, 0.02%; 8--40 min, 0.034%) and 210 mM KCI (lane 9--35 min, 0.02%; lane 10--70 min, 0.03%). The DNA products synthesized were analyzed on a 1.5% alkaline agarose gel as described under "Methods." The DNA size markers used were the same as those described in the legend to Figure 3. The smearing observed between the 1925 nt and 829/818 nt markers in lanes 5, 7 and 9 is an artefact and does not represent nascent DNA strands. 136 __ IO650 ‘_ 5386 _. 35I2 _ 2895 _ I925 l207 829, BIB 652 543 472 454 38| 357 272 lIIIIll I76 I56 I29 137 Processivity on Natural DNA and Influence of Single-stranded DNA Binding Protein (8 SB) Preliminary results indicated that the DNA synthetic mechanism of the near- homogeneous enzyme on a natural viral DNA, ch74, is complex. Kinetic measurements suggested that DNA polymerization proceeds in a quasi-processive manner. That is, the enzyme incorporates ~30 nucleotides per primer-terminus binding event. Although direct analysis revealed the presence of products of ~30 nucleotides, a substantial proportion of the synthetic products were actually multiples of this value (1 1). Thus, the enzyme has the capacity to remain associated with the primer-terminus and repeat its unit of synthesis several times prior to dissociation. Discrete products in multiples of ~20-50 nucleotides were also observed on singly- primed M13 DNA suggesting that this phenomenon is independent of the natural DNA template used and a characteristic of the DNA synthetic mechanism of the mitochondrial enzyme. Further, multiples of the processive unit are evident (Figure 5, lanes 1-3). At these low extents of synthesis the processivity on M13 DNA (Figure 5, lane 1) was nearly the same as that observed on singly-primed aXl74 DNA (11), and in fact, the average product size (~150 nucleotides) was the same as that obtained on the synthetic homopolymer poly (dA)-oligo (dT) (Figure 28). The rate of DNA synthesis by the mitochondrial DNA polymerase on the singly- primed M13 DNA was stimulated 3-fold by inclusion of E. 53211 SSB (Figure 6A). The average product size as determined by densitometric scanning of a 1.5% alkaline agarose gel (not shown) was increased by the same factor to approximately 400 nucleotides (Figure 6B) with the concurrent disappearance of the ~30 nt products as shown on a denatruing polyacrylamide gel (Figure 5, lanes 4-6). Thus, the heterologous SSB increased the rate of DNA synthesis and facilitated the generation of longer products by DNA polymerase yon natural DNA. This effect was also observed with the isolated 182,000 dalton subunit of 138 Figure 5. Analysis of the DNA products synthesized by DNA polymerase yon singly-primed M13 DNA and the effect of single-stranded DNA binding protein. The DNA products synthesized by DNA polymerase y were analyzed on a 6% polyacrylamide/7 M urea gel as described under "Methods." The reaction conditions and extents of DNA synthesis (%) obtained for each reaction after incubation for 5, 15 or 30 minutes were: No SSB (lanes 1-3)--0.04, 0.1 and 0.2; Plus SSB (lanes 4-6)--0.08, 0.32 and 0.6. The DNA size markers used were the same as those described in the legend to Figure 3 with the inclusion of a[32P]T'I'P end-labeled M13 trx 22 RF DNA (Eco RI: 10,650 nt) and a dideoxy sequencing ladder generated by incubation of the singly-primed M13 n 22 DNA with DNA polymerase I Klenow fiagment (Sanger et al., 1979; not shown). The product sizes indicated alphabetically are: a-20, b-52, c-85, d-95, e-l35, f-190, g-230, h- 280, i-310, j-440, k-500, 1-540 and m-600 nucleotides. a O 3-.-.m—3 I23456 8l8- I925 5 4 472 , 454 38] , 357 272 I76 I56 I29 I23 5] 140 Figure 6. Effect of a heterologous single-suanded DNA binding protein on the rate of DNA synthesis and processivity of DNA polymerase yon singly-primed M13 DNA. The rate of DNA synthesis was determined as described under Materials and Methods except that the reaction mix (0.175 ml) contained 32 LLM singly-primed M13 DNA; 30 1.1M each dGTP, dCTP, dTTP and 0t[32P]dATP (3700 cpm/pmol) and 0.] unit of DNA polymerase y fraction VI. E. 9111 SSB (2.5 ug) was included in an identical reaction mixtm'e and incubated on ice for 5 minutes prior to addition of enzyme (panel A). The weighted average size of the products of processive DNA synthesis with (0) and without (a) the presence of SSB were determined as described under "Methods" (panel B). 141 1 6 EXTENT OF SYNTHESIS, l B slim an. mile to mongoose: :5sz .8886 A L . _ 8 6 4 2 00 _o_ x as £8526 to ezmexm TIME, minutes °/. x 10' 142 Mia DNA polymerase at which was examined in a similar experiment (6), as well as the homologous enzyme, E. co_li DNA polymerase III (5). Effect of Template-primer and Enzyme Concentration on Processivity To detemrine if the template-primer concentration affects the mechanism of DNA polymerization we have determined the processivity under conditions where the ratio of primer ends to enzyme molecules was varied over an 8-fold range (Figure 7). This was necessary in order to determine if the high template-primer concentrations used in the processivity determinations promote nonproductive binding of the enzyme to single- stranded DNA template regions. If the DNA polymerase has a high affinity for single- stranded DNA it may bind these regions nonproductively after completion of one or a few I processive cycles which could result in apparent low processivity. In these experiments the singly-primed M13 DNA concentration was varied from 4 1.1M to 32 BM while the amount of enzyme added to each reaction was held constant. The reaction rates were linear, and nearly identical, under each condition throughout the time of incubation. This confirms that at each concentration the DNA was in excess--a condition required for processivity determination. The results indicate that primer concentration alone has no effect upon processivity in the range from 68 (lanes 1 and 2) - 550 (lanes 7 and 8) primer ends per enzyme molecule. In a companion experiment, the ratio of primer ends per enzyme molecule was again varied. In contrast to the previous experiment, this was accomplished by altering the amount of enzyme added to the reactions which contained a constant amount of DNA (Figure 8). This was done to examine the possibility that protein/protein (or protein/processivity factor) interactions might affect processivity. In all cases the reaction rates were linear with time, again indicating substrate excess. The experiment demonstrated that there is no effect upon product distribution caused by varying the enzyme concentration throughout a 60-fold range, from 50 (lanes 1 143 Figure 7. Effect of template-primer concentration on the DNA products synthesized by DNA polymerase y. The sin gly-primed M13 DNA concentration in the reaction mix (0.175 ml) and the extents of DNA synthesis (%) obtained after incubation for 5 or 15 minutes were: 4 11M (lane 1--0.004, 2--0.008), 8 uM (lane 3--0.002, 4-- 0.004), 16 BM (lane 5--0.001, 6--0.003) and 32 11M (lane 7--0.005, 8--0.01). dGTP, dCTP, dTTP and 0t[32P]dATP (18,000 cpm/pmol) were present at 5 11M and 0.026 unit of the fraction VI enzyme was used. The DNA products synthesized were analyzed on a 6% polyacrylamide/7 M urea gel as described under "Methods." The DNA size markers used were the same as those described in the legend to Figure 5. 144 —- 8|8—l925 mmmwm mm mm :2. __ __ ..N?t.3§.ii.ot .. ti... 1!... t |2345678 , .. . 5...... ..éfilsrrt ...: a li~n r .v1. m .333... r. , 2.91.»... .3; . 145 Figure 8. Effect of enzyme concentration on the DNA products synthesized by DNA polymerase y. The reaction mixture (0.175 ml) contained singly-primed M13 DNA (50 M) and 5 uM each dGTP, dCTP, dTTP and 0t[32P]dATP (25,000 cpm/pmol). The amount of DNA polymerase y fraction VI added and, in parentheses, the time of incubation and the extents of DNA synthesis (as percent) obtained in the reactions were as follows: 0.008 unit (lane 1--30 min, 0.004; 2--60 min, 0.007), 0.028 unit (lane 3--8 min, 0.004; 4--l6 min, 0.007), 0.112 unit (lane 5--2 min, 0.003; 6--4 min, 0.005) and 0.448 unit (lane 7--0.5 min, 0.004; 8--l min, 0.006). The DNA products synthesized were analyzed on a 6% polyacrylamide/7 M urea gel as described under "Methods." The DNA size markers used were the same as those described in the legend to Figure 5. I I II W .0 146 34 3:: w M-OI” 56 78 ..q. 60‘ o «can. — 8I8 - I925 #652 — 543 “472 —38I -—272 — I76 — I56 I29 I23 147 and 2) - 3000 (lanes 7 and 8) primer termini per enzyme molecule. Further, the product distribution was identical to that obtained when the template-primer concentration was varied (Figure 7). These results show clearly that the DNA template concentration has no influence on the processivity of DNA polymerase y. Further, there does not appear to be any dependence of enzyme concentration on the degree of processivity. This suggests that no higher order interactions occur in yigg which influence the template usage by the mitochondrial DNA polymerase. These results do not support the possibility that interactions occur either between intact enzyme molecules, or the intact enzyme and a dissociable polypeptide, or the intact enzyme and another unidentified factor present in the enzyme preparation. Comparison of the Processivity of Crude and Highly Purified DNA polymerase y In further efforts to identify positive or negative effectors of the mitochondrial DNA polymerase we utilized both less and more highly purified fractions of DNA polymerase y. This was done in an attempt to determine if these enzyme fractions contain dissociable factors that influence the polymerase processivity on natural DNA. The direct measurement of the processivity of the purified enzyme (fraction VI) in substrate excess was compared to that obtained with a crude enzyme fiaction (octyl-Sepharose, fraction IV; 9) and a mitochondrial DNA polymerase fraction VI more extensively purified by FPLC-gel filtration (fraction VII). Contaminating nuclease levels are very high in DNA polymerase y fiactions 1-111, and consequently, these enzyme fiactions were not examined. The crude enzyme (30% pure) generated a product distribution (Figure 9; lanes 1-3) identical to that of the near- homogeneous enzyme (lanes 4-6). The presence of contaminating nuclease in the crude fraction is evident; some products were degraded with increasing time of incubation (lane 3). No changes in product distribution were observed when the enzyme that was further chromatographed by gel filtration was analyzed (lanes 7 148 Figure 9. Comparison of the DNA products synthesized by other mitochondrial DNA polymerase yfiactions. The reaction mixtures (0.175 ml) contained 50 BM singly-primed M13 DNA, 30 11M each dATP, dGTP, dT'I'P and at[32P]dCTP (6000-12,000 cpm/pmol) and enzyme: octyl-Sepharose fraction IV (0.018 unit-- lanes 1-3: 0.02, 0.04 and 0.09% synthesis, respectively), glycerol gradient fraction VI (0.056 unit--lanes 4-6: 0.02, 0.04 and 0.06%) and FPLC-gel filtration purified fraction VII (0.046 unit-~1anes 7 and 8: 0.03 and 0.05%). The DNA products synthesized were analyzed on a 6% polyacrylamidefl M urea gel as described under "Methods." The DNA size markers used were the same as those described in the legend to Figure 5. 149 I 23 456 78 _ BIB-1925 #652 .. - :33 '— 381 I O r O I n. at 150 and 8). These results indicate that no dissociable processivity factors are present in the crude enzyme preparation (fraction IV) or removed from the highly purified enzyme by further chromatography which influence the product distribution generated by DNA polymerase y. I IN Reaction conditions such as template-primer composition, temperatme and ionic strength have long been known to have significant effects upon the processivity of E. goii DNA polymerase I (1). More recently, it was shown that other reaction conditions such as divalent cation concentration (3) and reaction pH affect the association of calf thymus DNA polymerases at and 8 with the template primer (2,4). We have shown that unlike calf thymus DNA polymerases at and 8, the low apparent processivity of 12mm DNA polymerase y is relatively unaffected by alterations in reaction pH. Further, it is clear that the processivity is not affected by the complexity of the template-primer. The value we observed was the same on the synthetic DNA template-primer and the natural DNAs which are known to possess considerable secondary structure. Remarkably, the mitochondrial DNA polymerase is capable of highly processive DNA synthesis when salt is entirely omitted from the DNA synthesis reactions. However, the physiological significance of this observation is not Clear. The infrastructtue of the rat liver mitochondrial matrix contains a high concenuation of protein molecules (~56% w/W; 12) but the ionic strength within the mitochondrial matrix is not known. These protein molecules and other metabolites (for example, salts and nucleotides) are thought to exist in a highly ordered state (12,13). Thus, the kinetic behavior of the mitochondrial DNA polymerase under Constrained natmal conditions is likely to be very different from that observed in solution. 151 Alterations in the ionic strength of the DNA synthesis reaction could affect the conformation of the template-primer, the DNA polymerase or both. Since DNA polymerase y has a relatively high affinity for predominantly single-stranded DNA template-primers under moderate concentrations of monovalent ion (1 1) it seems likely that the ionic interactions destabilize the association of the enzyme with the primer terminus during nucleotide incorporation--the elongation phase of DNA synthesis. The kinetic and product analysis data (11; this paper) suggest that DNA polymerase y is most susceptible to disruption after incorporation of the 30-50 nucleotides that constitute a processive synthetic cycle. This may indicate that the enzyme undergoes a salt sensitive conformational change after each blast of processive synthesis and releases the template-primer. Under reduced ionic strength conditions this forced disruption of the enzyme from the template-primer may be eliminated. In this case, even though the rate of DNA synthesis is reduced 7-fold, the enzyme has the opportunity to remain associated with the primer terminus after each processive cycle and to continue to replicate the available template producing full length copies. Notably, other well characterized DNA polymerases exhibit their highest processivity values at or near ionic conditions that are optimal for maximum synthetic rate. E. cgli DNA polymerase I (1) and Mia DNA polymerase (1 (6,14) are processive for ~20 nucleotides at low ionic strength. Increasing the salt concentration in the synthesis reactions converted E. 12911 DNA polymerase I to an entirely distributive mechanism of DNA synthesis (1). Raising the KC] concentration from 50 mM to 200 mM reduces the processivity of the calf thymus DNA polymerase-primase from 18 to 9 nucleotides (3) while the KB cell DNA polymerase B is unaffected by a 5-fold variation in KC] concentration (1). The highly purified chick embryo DNA polymerase y is highly processive on a poly (rA)-oligo (dT) template-primer under high salt conditions (15). Factors that may interact directly with the DNA template such as DNA binding proteins, may have important functions in allowing increased DNA polymerase 152 processivity. Like E. 9911 DNA polymerase III holoenzyme (5), the isolated 182,000 dalton subunit of mm DNA polymerase 01 (6), and Ma DNA polymerase- primase complex (3), the processivity of Dmsgphiia DNA polymerase y is increased by the addition of E. 9911 sin gle-stranded DNA binding protein. In the latter case, the increase in product size is in direct proportion to the increase in enzymatic reaction rate. In contrast, the processivity of the DNA polymerase-primase complex from calf thymus was inhibited by SSB (3). E. 9911 SSB may function 11; 39°99 to minimize template-primer secondary structure which may obstruct the progress of the enzyme. Alternatively, the SSB may direct the polymerase to the primer-terminus preventing or minimizing binding of the enzyme to single-suanded regions of the template. Inclusion of the heterologous SSB is the only positive protein effector of DNA polymerase y processivity that we have identified. Although we have not yet isolated such a protein from Mia, embryos, our data suggests that a homologous counterpart to E. QQIi SSB may be an important factor 19 m. In this regard, a mitochondrial single-stranded DNA binding protein purified fiom X99999: E9215 oocytes (16) was inhibitory to partially purified preparations of the homologous mitochondrial DNA polymerase, but the effect upon polymerase processivity was not determined. It is important to recognize that although SSB does increase the processivity of 121959911113 DNA polymerase y 2-3 fold it does not enable the enzyme to become completely processive. Further, it is clear that more complicated mechanisms of regulating DNA polymerase association with the template-primer may exist. Protein factors that enable the homologous DNA polymerase 01 to utilize denatured or sin gly-primed DNAs have been isolated from human HeLa cells (17), Ehrlich ascites tumor cells (18), monkey kidney (19) and mouse cell lines (20). These factors stimulated the synthetic rate of the DNA polymerase on predominantly single-standed templates to a level nearly equivalent to that observed using gapped DN As which are the preferred template-primers. The stimulatory 153 factors facilitate primer recognition by the DNA polymerase complex (19,20), thereby minimizing non-productive binding of DNA polymerase 01 to single-stranded template regions (21). However, these primer recognition proteins do not increase the polymerase processivity (19,20). The high processivity characteristic of E. 9911 DNA polymerase III holoenzyme is known to be related to the interaction of the core polymerase with its B subunit (7). Likewise, the processivity of the fetal calf thymus DNA polymerase 8 may be regulated by the proliferating cell nuclear antigen (PCNA; 22). Inclusion of PCNA increases the processivity of DNA polymerase 8 from < 30 nucleotides to > 200 nucleotides and is thought to allow the enzyme to function more efficiently on DNAs with low primer/template ratios. ' An 8-fold variation in template-primer concentration has no effect on the DNA polymerase y activity or processivity--suggesting that unlike DNA polymerase at (21) non- productive binding of the enzyme to single-stranded regions of the template-primer does not limit processivity. Moreover, it is also clear that the enzyme concentration has no effect. Even at very- low ratios of enzyme molecules to primer-termini the product distribution is unchanged. This implies that the enzyme molecule does not dissociate and rebind the primer-terminus after its incremental unit of synthesis but that it remains bound for nearly 150 nucleotides before dissociation occurs. It also indicates that no concentration dependent protein/protein interactions occm' which alter the template usage. This conclusion is reinforced by the fact that neither less nor more highly purified enzyme fractions exhibit any differences in processivity. This implies that there are no dissociable protein (or other) factors present in the crude enzyme preparations, or removed from the near-homogeneous enzyme fractions which influence the template-primer usage of DNA polymerase y. Consequently, no contaminating template recognition or processivity factors were detected in the enzyme fractions examined. 154 It is not known with certainty whether mitochondrial DNA is synthesized in a continuous or semidiscontinuous manner. The finding that the leading DNA strand initiates at a single origin of mitochondrial DNA replication suggests that its synthesis is most likely continuous. The lagging strand origin has not been identified in Dr9gophila, but a comparable region, known as the light (L)-strand origin, has been mapped in the human, bovine and mouse mtDNAs (23). These L-strand origins have the potential to form a specific secondary structure once the parental H-strand has been displaced during DNA synthesis. This region is the site of DNA primer synthesis in mouse and human mtDNAs. Once the primer is synthesized it is thought to be extended by the mitochondrial DNA polymerase in a continuous manner. However, it is not known whether L-strand replication can initiate at other regions of the H-strand lacking the ability to form this particular secondary structtu'e. Electron microscopic studies have revealed the extreme asymmetric nature of mm mitochondrial DNA replication (24). Because the leading DNA strand can be almost completely replicated before lagging DNA strand synthesis initiates (8) the mitochondrial DNA polymerase must have the ability to efficiently function on DNA templates with widely varying degrees of single-strandedness. Thus, the mitochondrial enzyme encounters an entirely double-stranded template during leading strand synthesis; at the time of initiation of DNA synthesis on the displaced DNA strand, the template is predominantly or entirely single-stranded (8). There is no data to suggest that two distinct DNA polymerases are required to replicate the complex DNA template that is presented by the mitochondrial genome. Therefore, the mitochondrial DNA polymerase must have the capacity to accomplish this task itself or in concert with other unidentified accessory factors. Certainly, the Whila mitochondrial DNA polymerase in its current state of purification is not highly processive under reaction conditions that are optimal for synthetic rate on homopolymeric and natural DNAs in yin, and its template usage properties set it 155 apart from any other DNA polymerases fiom both prokaryotic and eukaryotic sources. Whether or not the mitochondrial DNA polymerase must be highly processive iii, £in is Still in question. In fact, we have calculated that the enzyme is present in moderate excess (30-40-fold) over mitochondrial DNA molecules in 2129. Our previous studies indicate that the enzyme can efficiently replicate predominantly sin gle-stranded DNAs when present in moderate excess (1 l). 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. W Bambara, R.A., Uyemura, D. and Choi, T. (1978) J. Biol. Chem. 253: 413-423. Tan, C.-K., So, M.J., Downey, K.M. and 80, AG. (1987) Nucl. Acids Res. 15: 2269-2278. Hohn, K.-T. and Grosse, F. (1987) Biochemistry 26: 2870-2878. Sabatino, R.D., Myers, T.W., Bambara, R.A., Kwon-Shin, 0., Marraccino, R.L. and Frickey, PH. (1988) Biochemistry 27 : 2998-3004. Fay, P.J., Johanson, K.O., McHenry, CS. and Bambara, R.A. (1981) J. Biol.Chem. 258: 11344-11349. Cotterill, S., Chui, G. and Lehman, LR. (1987) J. Biol. Chem. 262:16100- 16104. LaDuca, R.J., Crute, J.J., McHenry, CS. and Bambara, R.A. (1986) J. Biol. Chem. 261: 7550-7557. 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. Wemette, CM. and Kaguni, LS (1986) J. Biol. Chem. 261: 14764-14770. Kaguni, L.S., Wemette, C.M., Conway, M.C. and Yang-Cashman, P. (1988) In: Cancer Cells: Eukaryotic DNA Replication, Vol. 6, (TJ. Kelly and B.W. Stillman, eds.) Cold Spring Harbor Laboratory, Cold Spring Harbor, New York Wemette, C.M., Conway, M.C. and Kaguni, LS. (1988) Biochemistry 27 : 6046-6054.. Srere, RA. (1980) Trends Biochem. Sci. 5: 120-121. Srere, RA. (1981) Trends Biochem. Sci. 6: 4-7. Villani, G., Fay, P.J., Bambara, R.A. and Lehman, LR. (1981) J. Biol. Chem. 256: 8202-8207. Yamaguchi, M., Matsukage, A. and Takahashi, T. (1980) Nature 285:45-47. Mignotte, B., Barat, M. and Mounolou, J.-C. (1985) Nucl. Acids Res. 13: 1703- 1716. Lamothe, P., Baril, B., Chi, A., Lee, L. and Baril E. (1981) Proc. Natl. Acad. Sci. U.S.A . 78: 4723-4727. Faust, EA. and Rankin, CD. (1982) Nucl. Acids Res. 10: 4181-4201. _ Pritchard, CG. and dePamphilis, M.C. (1983) J. Biol. Chem. 258: 9801-9809. 156 20. 21. 22. 23. 24. 157 Kawasaki, K., Nagata, K., Enomoto, T., Hanaoka, F. and Yamada, M. (1984) J. Biochem 95: 485-493. Fisher, P.A., Chen, T.T. and Korn, D. (198]) J. Biol. Chem. 256: 133-141. Prelich, G., Tan, C.-K., Kostura, M., Matthews, M.B., So, A.G., Downey, K.M. and Stillman, B. (1987) Nature 326: 517-520. Wong, T.W. and Clayton, DA. (1985) Cell 42: 951-958. Goddard, J .M. and Wolstenholme, DR. (1978) Proc. Natl. Acad. Sci. USA. 75: 3886—3890. Chapter V Summary and Perspectives 158 159 Mitochondrial DNA polymerase was purified 2500-fold from crude mitochondria derived from Dr9soghila m9lanoga§t9r embryos. The purification involves successive chromatography of a salt/detergent mitochondrial extract on phosphocellulose, single- stranded DNA cellulose, octyl-Sepharose and Cibacron blue agarose. Velocity sedimentation of the enzyme on a 10-30% glycerol gradient is utilized as the final purification step; the entire procedm'e is completed in seven days. Starting with a ' mitochondrial extract prepared from 200 grams of 129959911114 embryos this procedure yields ~10 ug of near-homogeneous DNA polymerase y. Inclusion of Triton X- 100 in the elution buffers used in steps IV -VI of the purification is essential for recovery of enzyme activity. If the detergent is omitted no DNA polymerase activity is detectable. This indicates that the enzyme is probably very hydrophobic in nature, and may suggest the importance of a membrane association for enzymatic function in mg. This study represents the first reported purification to near-homogeneity of a mitochondrial DNA polymerase from Ma and it has provided a number of new findings that have clarified and expanded our understanding of the mitochondrial enzyme and its function in 21179. First, these studies have delineated the physical structure of the enzyme. The 2199911113 mitochondrial DNA polymerase is a heterodimer comprised of one 125,000 dalton polypeptide and one 35,000 dalton polypeptide. The chick embryo DNA polymerase y (the only other highly pmified mitochondrial DNA polymerase) is reported to be a homotetramer composed of four 47 ,000 dalton subunits (Literature Review, p. 23). While this may be the case, it has recently become apparent that another polypeptide of ~135,0(X) daltons may be important for function of the chick embryo mitochondrial DNA polymerase. Consequently, the actual status of the subunit structure of the chick embryo enzyme remains to be resolved. While DNA polymerase y as isolated fiom the two sources may or may not share similar subunit structures, my W studies were the first to examine DNA polymerization in 51111. This is the first report to directly identify a mitochondrial enzyme 160 subunit responsible for DNA polymerization. The 125,000 dalton subunit provides this function in both the crude and highly purified forms of the enzyme. Comparison of i_Ii 5119 DNA polymerization of both the crude and highly purified enzyme also provides an indication that it has not undergone extensive proteolysis during its purification. Proteolysis has been a severe problem in the purification of other DNA polymerases. The importance of the 125,000 dalton subunit is established. However, it is not at all clear what function the 35,000 dalton subunit provides. It is possible that this subunit regulates the efficiency of nucleotide polymerization by some unidentified mechanism. It may affect template-primer usage by increasing (or decreasing) the affinity of the DNA polymerase for the primer terminus. Alternatively, it may play a role in regulating the processivity or replication fidelity of the enzyme. In its crurent state, the Mill]; mitochondrial DNA polymerase does not exhibit any DNA primase, DNA-dependent ATPase or strand displacement activities. Therefore, it seems unlikely that the 35,000 dalton subunit is involved in any of these functions. Enzyme subunit dissociation studies could be utilized to provide information regarding the function of the 35,000 dalton subunit of DNA polymerase y. The isolated subunits, as well as the reconstituted enzyme could be evaluated to determine optimal reaction conditions (monovalent or divalent ion, pH) and kinetic parameters (Km for DNA or dNTPs). Comparison of the results to those presented in this study utilizing the intact DNA polymerase ymay provide an indication of the influence of the small subunit on catalytic activity. In addition the isolated enzyme subunits could be assessed for the presence of other enzymatic activities which may be required for mtDNA replication (for example: the DNA primase, DNA-dependent ATPase or strand displacement activities which were mentioned above). It is possible that separation of the DNA polymerase subunits may unmask the presence of an enzymatic activity that is not detectable in the intact enzyme. In this regard, a potent 3'-5' exonuclease activity associated with the isolated 182,000 dalton subunit of Win DNA polymerase (I is detectable. However, 161 no exonuclease activity is apparent when the intact multi-subunit enzyme is examined in the same assay (Literature Review, p. 28). This study has also demonstrated that the 12195521211114 DNA polymerase y exhibits a high degree of accuracy in DNA replication. This is in agreement with the studies of the chick embryo mitochondrial DNA polymerase (Literature Review, p. 30). Results fiom our laboratory indicate that like the chick embryo enzyme the 121939913119 mitochondrial DNA polymerase possesses a 3'-5' exonuclease activity which may contribute to replication fidelity. Examination of the fidelity of DNA replication of the isolated 125,000 dalton subunit may provide an indication that the 35,000 dalton subunit has an effect on the accuracy of DNA polymerization in 21932. Also, these studies may allow assignment of the 3'-5' exonuclease activity to a particular enzyme subunit. Unlike other eukaryotic DNA polymerases, the intact mitochondrial DNA polymerase efficiently utilizes a variety of template-primers. Template-primer utilization may be modified by the presence or absence of the small enzyme subunit. The natural and synthetic template-primers examined in this study vary in nucleotide sequence complexity, primer density and degree of single-strandedness. Under optimal reaction conditions and moderate enzyme excess the intact DNA polymerase y can completely copy the available template. This is accomplished even though the enzyme is sensitive to template secondary structure, a finding which has not been reported previously. In addition, the unusual quasi-processive nature of the enzyme, determined under conditions of substrate excess, is unlike that of any other DNA polymerase from prokaryotic or eukaryotic sources. The processivity of DNA polymerase y is nearly identical when examined on several template- primers or under a variety of reaction conditions. Under optimal reaction conditions the 12mm enzyme generates relatively long products that consist of multiples of the increment of processive synthesis but these products do not attain full length. No factors which influence template usage were detected in the enzyme fractions examined. These studies have also demonstrated for the first time that the processivity of a mitochondrial 162 DNA polymerase can be increased 2-3-fold by inclusion of a single-stranded DNA binding protein. Further, Mia DNA polymerase ycan become completely processive under low ionic strength conditions which allow the enzyme to remain associated with the template-primer although the reaction rate is reduced 7-8-fold. Further efforts to examine the mitochondrial DNA polymerase will require the use of antibodies specific for the intact enzyme and its isolated subunits. Generation of monoclonal antibodies will allow development of a revised purification scheme for the mitochondrial DNA polymerase. It is anticipated that the use of an affinity purification step utilizing a monoclonal antibody affinity column will result in a higher yield of DNA polymerase y. This will be useful because it will provide more material for the structural and functional analysis of the enzyme. Such studies may include peptide mapping and localization of the DNA and dNTP binding sites. Also, it may be possible to isolate enough of each of the isolated subunits to obtain partial nucleotide sequences. This information could be utilized to generate oligonucleotide probes for use in cloning the genes encoding the the mitochondrial DNA polymerase subunits. In addition, a more rapid purification scheme using an affinity column may allow isolation of the mitochondrial DNA polymerase associated with other polypeptides which possess important functions necessary for mitochondrial DNA replication. These studies may identify further alterations in catalytic activity and efficiency. Also, they may result in redefinition of the holoenzyme structrue of DNA polymerase y. Finally, fluorescent labeling techniques using antibodies specific for DNA polymerase y will allow microscopic examination of the cellular location of the enzyme. These studies could provide direct evidence which may substantiate or refute earlier reports that a small amount of DNA polymerase y is localized within the isolated cell nucleus (Literature Review, p. 21). Positive identification of the presence of DNA polymerase y within the nucleus could indicate that the enzyme provides an important function within that 163 cellular compartment. Subsequent studies could elucidate a replicative function for DNA polymerase y. Our studies have indicated that both salt and detergent are required for extraction of DNA polymerase y from crude mitochondria. Other results (not presented) indicate that the enzyme can be isolated from mitochondria that have been treated with digitonin; a detergent that removes the outer mitochondrial membrane. Further, the presence of detergent is required to maintain the activity of the W mitochondrial DNA polymerase in 31199. Taken together, the data suggest that the enzyme is located within the mitochondrial matrix 19 _v_iyg; perhaps in association with the inner mitochondrial membrane. Direct analysis of the location of DNA polymerase y within the mitochondrion will contribute important information about its organization and function in mg. The studies presented in this report represent the most extensive examination of the physical and catalytic properties of a single highly purified mitochondrial DNA polymerase. Further efforts to examine the enzymatic mechanism of mitochondrial DNA replication will require isolation and Characterization of other mitochondrial replication proteins. These might include a mitochondrial RNA polymerase or DNA primase which may be necessary to synthesize primers utilized by the DNA polymerase, a DNA helicase to unwind duplex DNA strands and a mitochondrial DNA binding protein to protect DNA strands from nucleases and/or facilitate the DNA polymerase by minimizing template secondary structure. Once these and other components are isolated from a single source their interaction with the mitochondrial DNA polymerase can be examined and the reactions required for initiation of mitochondrial DNA synthesis can be reconstituted in m. APPENDIX A 164 Determination of native molecular mass of mm DNA polymerase y M=6ltnNas/(l-vp) n = viscosity of H20 (0.01 g cm‘1 3‘1) 8 = sedimentation coefficient (7.6 x 10'13 s) N = Avogadros number (6.3 x 10‘23 mole'l) v = partial specific volume (0.725 cm3 g‘l) p = density of H20 (1 g cm‘3) a = Stokes radius (51.2 x 10‘8 cm) M = molecular mass (g mole'l) = 168,000 g mole'l Reference: Siege], L. M. and Monty, K. J. (1966) Biochim. Biophys. Acta. 112: 346- 362. 165 APPENDIX B 166 PUBLICATIONS Wemette, C. M., SanClemente, C. L. and Kabara, J. J. (1981) The effect of surfactants upon the activity and distribution of glucosyltransferase in W mgtang 6715. Pharmacology and Therapeutics in Dentistry. 6: 99-107. Kabara, J. J. and Wemette, C. M. (1982) Cosmetic formulas preserved with food-grade chemicals-Part 11. Cosmetics and Toiletries 97: 77-84. Wernette, C. M. and Kaguni, L. S. (1986) A mitochondrial DNA polymerase from embryos of Mg W purification, subunit structure and partial characterization. J. Biol. Chem. 261: 14764- 14770. Su, C.-J., da Cunha, A., Wemette, C. M., Reusch, R. N. and Sadoff, H. L. (1987) Protein synthesis during encystrnent of W 39' n9199giii. .L Bag. 169: 4451-4456. Kaguni, L. S., Wemette, C. M., Conway, M. C. and Yang-Cashman, P. (1988) Structural and catalytic features of the mitochondrial DNA pol from Drggophila W embryos. In: Cancer Cells: Eukaryotic DNA Replication,Vol. 6 (T. J. Kelly and B. W. Stillman, eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, New York . Wemette, C. M., Conway, M. C. and Kaguni, L. S. (1988) The mitochondrial DNA polymerase from 211299911113 W embryos: kinetics, processivity and fidelity of DNA polymerization. Biochemistry 27 : 6046-6054 Wemette, C. M. and Kaguni, L. S. (1988) Processivity of the mitochondrial DNA polymerase from Mia embryos: template-primer, pH and DNA binding protein effects (in preparation). 167