Mammalian DNA polymerase a holoenzymes with possible functions at the leading and lagging strand of the replication fork ( DEAE-cellulose chromatography / primase / DNA-dependent ATPase / exonuclease / DNA topoisomerase )
H. Ottiger,U. Hubscher
Abstract:At an early purification stage, DNA polymerase a holoenzyme from calf thymus can be separated into four different forms by chromatography on DEAE-cellulose. All four enzyme forms (termed A, B, C, and D) are capable of replicating long single-stranded DNA templates, such as parvoviral DNA or primed M13 DNA. Peak A possesses, in addition to the DNA polymerase a, a double-stranded DNA-dependent ATPase, as well as DNA topoisomerase type II, 3'-5' exonuclease, and RNase H activity. Peaks B, C, and D all contain, together with DNA polymerase a, activities of primase and DNA topoisomerase type II. Furthermore, peak B is enriched in an RNase H, and peaks C and D are enriched in a 3-5' exonuclease. DNA methylase (DNA methyltransferase) was preferentially identified in peaks C and D. Velocity sedimentation analyses of the four peaks gave evidence of unexpectedly large forms of DNA polymerase a (>11.3 s), indicating that copurification of the above putative replication enzymes is not fortuitous. With moderate and high concentrations of salt, enzyme activities cosedimented with DNA polymerase a. Peak C is more resistant to inhibition by salt and spermidine than the other three enzyme forms. These results suggest the existence of a leading strand replicase (peak A) and several lagging strand replicase forms (peaks B, C, and D). Finally, the saltresistant C form might represent a functional DNA polymerase a holoenzyme, possibly fitting in a higher-order structure, such as the replisome or even the chromatin. DNA replication requires the concerted action of many proteins and enzymes either separately or in a complex (1, 2). These include, for instance, DNA polymerases, DNA polymerase accessory proteins, primase, topoisomerase, helicase, DNA-binding proteins, RNase, and ligase (1, 2). Studies on prokaryotic in vitro DNA replication using small bacteriophages or plasmids containing the oriC sequence as model replicons, show that >20 polypeptides have to operate coordinately (1-3). In vitro adenovirus DNA replication, representing the best studied eukaryotic DNA replication model system so far, again attests to the complex interaction of many proteins (e.g., see ref. 4). The term replisome has been introduced to denote the hypothetical structure working at the prokaryotic replication fork (2). It may consist of a conserved primosome (5), DNA polymerase III holoenzyme (6), helicase, as well as topoisomerase. Escherichia coli DNA polymerase III holoenzyme is essentially responsible for prokaryotic chromosomal replication (6). Detailed knowledge of this functional multipolypeptide complex has accumulated in the last few years (for review, see ref. 7). In analogy to the prokaryotic DNA polymerase III holoenzyme, we recently described a similar form in eukaryotes (8, 9). Using defined in vivo-like templates, we isolated a complex form of DNA polymerase a from freshly harvested calf thymus (8, 9). This enzyme, termed the holoenzyme, is functionally and physicochemically distinct from a homogeneous DNA polymerase a, termed the core enzyme (8, 10), isolated from the same source. In the course of these studies, we realized that DNA polymerase a holoenzyme extracted from calf thymus and partially purified by phosphocellulose chromatography reproducibly splits into four different DNA polymerase a forms on chromatography on DEAE-cellulose. We provide evidence that at least six additional enzyme activities, whose involvement in DNA replication must be postulated, copurify with one or the other of these four DNA polymerase forms. The results suggest that the eukaryotic DNA polymerase a holoenzyme can be separated into forms that could function at the leading and at the lagging strand. MATERIALS AND METHODS Chemicals. Unlabeled NTPs, dNTPs, and araCTP were purchased from P-L Biochemicals. [3H]dTTP, [8-3H]ATP, [y32P]ATP, poly[8-3H]A, and S-adenosyl-L-[methyl-3H]methionine were from Amersham; pepstatin, novobiocin, and spermidine were from Sigma. Whatman was the supplier of DEAE-cellulose, and aphidicolin was provided by S. Spadari (Pavia, Italy). Production, Isolation, and Preparation of Nucleic Acids. Single-stranded (ss) DNA from minute virus of mice, a parvovirus, and phage M13 (mp8) were isolated as described (8). Activation and denaturation of calf thymus DNA and hybridization of poly(dA) to poly(dT)12.18 were done as described (8). Plasmid pj3G forms I and II were gifts from E. Di Capua (Zurich). Hemimethylated Micrococcus luteus DNA and supercoiled pAT153 DNA (both prepared according to ref. 11) were provided by S. Spadari (Pavia, Italy). Poly(dT) and poly(dA) oligo(dT) were purchased from P-L Biochemicals and Boehringer Mannheim, respectively. The hook polymer, po1y(dA)1j0-oligo(dT)25, was prepared as outlined by Fisher et al. (12). Primed M13 DNA was prepared as follows: the pentadecamer oligonucleotide T-C-C-C-A-G-TC-A-C-G-A-C-G-T (New England Biolabs) was hybridized to M13 ss DNA (mp8) in 10 mM Tris'HCl, pH 7.5/1 mM EDTA/30 mM NaCl, at a ratio of 1:400 (wt/wt). After incubation at 55°C for 5 min, the primed M13 DNA was cooled to room temperature and stored at 4°C until use. In Vitro Assays of DNA Synthesis. The reaction mixtures (25 ul) included the following components; (i) for activated and heat-denatured DNA: 20 mM potassium phosphate, pH 7.2/0.1 mM EDTA/4 mM dithiothreitol/bovine serum albumin (0.25 mg/ml)/10 mM MgCl2/dATP, dGTP, and dCTP each at 48 ,uM/18 ,uM [3H]dTTP (100-200 cpm/pmol)/16.5 nmol of activated (or heat-denatured) DNA as nucleotides and DNA polymerase to be assayed; incubation was at 37°C for 10 min; (ii) for poly(dA) oligo(dT)12_18 (different base ratios) and hook poly(dA)j10-oligo(dT)25: 20 mM Tris HCl, pH Abbreviations: ss, single stranded; ds, double stranded. 3993 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 3994 Biochemistry: Ottiger and Hubscher 7.5/4% (wt/vol) sucrose/8 mM dithiothreitol/bovine serum albumin (80 gg/ml)/1 mM MnCl2/18 ,uM [3H]dTTP (100-200 cpm/pmol)/225 pmol of homopolymer (as nucleotides) and enzymes to be assayed; incubation was at 370C for 60 min; (iii) for elongation of parvoviral ss DNA and primed M13 DNA: 20 mM Tris HCl, pH 7.5/4% (wt/vol) sucrose/8 mM dithiothreitol/bovine serum albumin (80 ,ug/ml)/10 mM MgCl2/5 mM ATP/10 mM KCl (omitted in M13 assay)/ dATP, dGTP, and dCTP each at 48 ,M/18 ,tM [3H]dTTP (100-200 cpm/pmol)/200 pmol (total nucleotides) of ss parvoviral or primed M13 DNA and enzyme to be assayed; incubation was at 37°C for 1 hr. When double-stranded (ds) pPG form I or II was used for assay, the reaction mixtures contained, in addition to the ingredients of the elongation assays, UTP, GTP, and CTP each at 100 ,uM. The reactions were stopped, the DNA was precipitated, the precipitate was collected on a Whatman GF/C filter and dried, and the radioactivity was determined as described (13). One unit of activity is defined as the incorporation of 1 nmol of dNTP into acid-insoluble DNA in 60 min at 37°C. Other Enzyme Assays. (i) Primase. The reaction was performed as described by Conway and Lehman by using 0.5 unit of DNA polymerase I per assay (14). One unit of primase activity is the amount that catalyzes the incorporation of 1 nmol of dATP (with M13 ss DNA of dNTPs) into acidinsoluble material in 60 min at 37°C. (ii) DNA-dependent ATPase. This assay was carried out exactly as described by Shlomai and Kornberg (15) by using denatured calf thymus DNA, 4X174 ss DNA or ds supercoiled pAT153 DNA at 6600, 400, and 600 pmol as nucleotides, respectively. One unit of DNA-dependent ATPase is defined as the cleavage of 1 nmol of ATP per 30 min at 37°C. (iii) 3'-5' exonuclease. The poly(dA-dT) substrate was labeled at the 3' terminus with [3H]dTTP, and the 3'-5' exonuclease activity was determined according to Byrnes et al. (16). One unit of 3'-5' exonuclease was determined as the conversion of 1 nmol of terminally labeled poly(dA-dT) [3H]dTMP into acid-soluble nucleotides in 60 min at 37°C. (iv) RNase H. The assay for RNase H and preparation of a RNA-DNA hybrid was as described by Berkower et al. (17). One unit of RNase H activity was the amount of enzyme producing 1 nmol of acid-soluble material in 30 min at 37°C. The control for RNase contamination was as described for RNase H except the RNA-DNA hybrid was replaced by 2.5 pmol of [3H]poly(A) (15,000 cpm/pmol). (v) DNA ligase. Ligation was measured according to Zimmermann and Levin (18). One unit of ligase activity is the amount that converts 1 pmol of 32p into a phosphatase-resistant form in 30 min at 37°C. E