Discontinuous DNA replication in vitro. I. Two distinct size classes of intermediates

A Serendipitous Path to Pharmacology

My path to research in neuropharmacology has been a coalescing of my training as a molecular biologist and my intense interest in an esoteric group of animals, the fish-hunting cone snails. Attempting to bridge these two disparate worlds has led me to an idiosyncratic career as a pharmacologist.

Repriming DNA synthesis: an intrinsic restart pathway that maintains efficient genome replication

To bypass a diverse range of fork stalling impediments encountered during genome replication, cells possess a variety of DNA damage tolerance (DDT) mechanisms including translesion synthesis, template switching, and fork reversal. These pathways function to bypass obstacles and allow efficient DNA synthesis to be maintained. In addition, lagging strand obstacles can also be circumvented by downstream priming during Okazaki fragment generation, leaving gaps to be filled post-replication. Whether repriming occurs on the leading strand has been intensely debated over the past half-century. Early studies indicated that both DNA strands were synthesised discontinuously. Although later studies suggested that leading strand synthesis was continuous, leading to the preferred semi-discontinuous replication model. However, more recently it has been established that replicative primases can perform leading strand repriming in prokaryotes. An analogous fork restart mechanism has also been identified in most eukaryotes, which possess a specialist primase called PrimPol that conducts repriming downstream of stalling lesions and structures. PrimPol also plays a more general role in maintaining efficient fork progression. Here, we review and discuss the historical evidence and recent discoveries that substantiate repriming as an intrinsic replication restart pathway for maintaining efficient genome duplication across all domains of life.

Historical Perspective of Eukaryotic DNA Replication

The replication of the genome of a eukaryotic cell is a complex process requiring the ordered assembly of multiprotein replisomes at many chromosomal sites. The process is strictly controlled during the cell cycle to ensure the complete and faithful transmission of genetic information to progeny cells. Our current understanding of the mechanisms of eukaryotic DNA replication has evolved over a period of more than 30 years through the efforts of many investigators. The aim of this perspective is to provide a brief history of the major advances during this period.

Discontinuous leading-strand synthesis: a stop-start story

Reconstitution experiments using replication proteins from a number of different model organisms have firmly established that, in vitro, DNA replication is semi-discontinuous: continuous on the leading strand and discontinuous on the lagging strand. The mechanism by which DNA is replicated in vivo is less clear. In fact, there have been many observations of discontinuous replication in the absence of exogenous DNA-damaging agents. It has also been proposed that replication is discontinuous on the leading strand at least in part because of DNA lesion bypass. Several recent studies have revealed mechanistic details of pathways where replication of the leading strand introduces discontinuities. These mechanisms and their potential contributions to observations of discontinuous replication in vivo will be discussed.

Next-generation sequencing of Okazaki fragments extracted from Saccharomyces cerevisiae

Genome-wide Okazaki fragment distribution can differentiate the discontinuous from the semi-discontinuous DNA replication model. Here, we investigated the genome-wide Okazaki fragment distribution in Saccharomyces cerevisiae S288C. We improved the method based upon lambda exonuclease digestion to purify Okazaki fragments from S288C yeast cells, followed by Illumina sequencing. The distribution of Okazaki fragments around confirmed replication origins, including two highly efficient replication origins, supported the discontinuous DNA replication model. In S. cerevisiae mitochondria, Okazaki fragments were overrepresented in the transcribed regions, indicating the interplay between transcription and DNA replication.

The replication intermediates in Escherichia coli are not the product of DNA processing or uracil excision

The current model of DNA replication in Escherichia coli postulates continuous synthesis of the leading strand, based on in vitro experiments with purified enzymes. In contrast, in vivo experiments in E. coli and its bacteriophages, in which maturation of replication intermediates was blocked, report discontinuous DNA synthesis of both the lagging and the leading strands. To address this discrepancy, we analyzed nascent DNA species from ThyA+ E. coli cells replicating their DNA in ligase-deficient conditions to block maturation of replication intermediates. We report here that the bulk of the newly synthesized DNA isolated from ligase-deficient cells have a length between 0.3 and 3 kb, with a minor fraction being longer that 11 kb but shorter than the chromosome. The low molecular weight of the replication intermediates is unchanged by blocking linear DNA processing with a recBCD mutation or by blocking uracil excision with an ung mutation. These results are consistent with the previously proposed discontinuous replication of the leading strand in E. coli.

Discontinuous or semi-discontinuous DNA replication in Escherichia coli?

The postulate that a stalled/collapsed replication fork will be generated when the replication complex encounters a UV-induced lesion in the template for leading-strand DNA synthesis is based on the model of semi-discontinuous DNA replication. A review of existing data indicates that the semi-discontinuous DNA replication model is supported by data from in vitro studies, while the discontinuous DNA replication model is supported by in vivo studies in Escherichia coli. Until the question of whether DNA replicates discontinuously in one or both strands is clearly resolved, any model building based on either one of the two DNA replication models should be treated with caution.

Ribosylnicotinamide kinase domain of NadR protein: identification and implications in NAD biosynthesis

NAD is an indispensable redox cofactor in all organisms. Most of the genes required for NAD biosynthesis in various species are known. Ribosylnicotinamide kinase (RNK) was among the few unknown (missing) genes involved with NAD salvage and recycling pathways. Using a comparative genome analysis involving reconstruction of NAD metabolism from genomic data, we predicted and experimentally verified that bacterial RNK is encoded within the 3' region of the nadR gene. Based on these results and previous data, the full-size multifunctional NadR protein (as in Escherichia coli) is composed of (i) an N-terminal DNA-binding domain involved in the transcriptional regulation of NAD biosynthesis, (ii) a central nicotinamide mononucleotide adenylyltransferase (NMNAT) domain, and (iii) a C-terminal RNK domain. The RNK and NMNAT enzymatic activities of recombinant NadR proteins from Salmonella enterica serovar Typhimurium and Haemophilus influenzae were quantitatively characterized. We propose a model for the complete salvage pathway from exogenous N-ribosylnicotinamide to NAD which involves the concerted action of the PnuC transporter and NRK, followed by the NMNAT activity of the NadR protein. Both the pnuC and nadR genes were proven to be essential for the growth and survival of H. influenzae, thus implicating them as potential narrow-spectrum drug targets.

The Escherichia coli NadR regulator is endowed with nicotinamide mononucleotide adenylyltransferase activity

The first identification and characterization of a catalytic activity associated with NadR protein is reported. A computer-aided search for sequence similarity revealed the presence in NadR of a 29-residue region highly conserved among known nicotinamide mononucleotide adenylyltransferases. The Escherichia coli nadR gene was cloned into a T7-based vector and overexpressed. In addition to functionally specific DNA binding properties, the homogeneous recombinant protein catalyzes NAD synthesis from nicotinamide mononucleotide and ATP.

Activity of the nicotinamide mononucleotide transport system is regulated in Salmonella typhimurium

Transport of nicotinamide mononucleotide (NMN) requires two functions, NadI(T) and PnuC. The PnuC protein is membrane associated, as judged by isolation of active TnphoA gene fusions and demonstration that the fusion protein is membrane associated. The PnuC function appears to be the major component of the transport system, since mutant alleles of the pnuC gene permit NMN transport in the absence of NadI(T) function. We present evidence that the activity of the NMN transport system varies in response to internal pyridine levels (presumably NAD). This control mechanism requires NadI(T) function, which is provided by a bifunctional protein encoded by the nadI gene (called nadR by Foster and co-workers [J. W. Foster, Y. K. Park, T. Fenger, and M. P. Spector, J. Bacteriol. 172:4187-4196]). The nadI protein regulates transcription of the nadA and nadB biosynthetic genes and modulates activity of the NMN permease; both regulatory activities respond to the internal pyridine nucleotide level.

Discontinuous DNA replication in a lig-7 strain of Escherichia coli is not the result of mismatch repair, nucleotide-excision repair, or the base-excision repair of DNA uracil

After pulse-labeling with 3H-thymidine for 30 s at 42 degrees C, the newly-synthesized DNA from uvrB5 lig-7, uvrB5 lig-7 ung-1 (or ung152), uvrB5 lig-7 mutL218 (or mutS215), and uvrB5 lig-7 ung-1 mutL218 (or mutS215) cells sedimented very slowly in alkaline sucrose gradients. The bulk of these DNA molecules were smaller than 2,000 nucleotides long (i.e., about the size of Okazaki fragments), and none of the 3H-radioactivity was found to sediment as high-molecular-weight DNA. These results indicate that the apparent discontinuous DNA replication observed in lig-7 strains is not the result of mismatch repair, nucleotide-excision repair, or the base-excision repair of DNA uracil.

Flexibility in RNA priming of Okazaki pieces at the E. coli replication fork

We present results which suggest considerable flexibility in the RNA priming of Okazaki pieces at the E. coli replication fork. Using film lysates on cellophane discs, we have identified RNA at the 5' ends of Okazaki pieces. All four ribonucleotides are found to be present at the RNA-DNA junction if all four ribonucleoside triphosphates are used. However, if only ATP, or ATP and GTP are used, then only 2' (3')AMP, or 2' (3')AMP and 2' (3')GMP are found at the RNA-DNA junction. A nearest neighbor analysis of RNA associated with Okazaki pieces using alpha 32P-CTP as a probe shows a similar dependence of nearest neighbor composition on the ribonucleoside triphosphate composition of the incubation mixture. Thus, the nucleotide composition of the RNA primers at the ends of Okazaki pieces varies as a function of the ribonucleoside triphosphates available.

DNA replication intermediates synthesized by lysates of dnaB, dnaG and dnaB dnaG mutants in vitro

Isogenic dnaB, dnaG, and dnaB dnaG mutants were constructed and used as extracts in the cellophane-disc in vitro DNA replication system. The increased proportion of 5S DNA characteristics of the dnab extract and the lack of Okazaki piece synthesis characteristic of the dnaG extract were both apparent in analysis of the dnaB dnaG mutant extract reaction. A hypothetical scheme to explain these results and those of others is presented.

In vitro replication of a DNA fragment containing the vicinity of the origin of E. coli DNA replication

The restriction nuclease cleavage pattern of E. coli DNA synthesized in vitro in the cellophane membrane system (Schaller et al., 1972) is similar to the one obtained after labelling E. coli in vivo. This is shown for exponentially growing cells and for cells synchronized by amino acid starvation followed by thymine starvation. In synchronized cells a piece of some 180 kilobase pairs is labelled containing oriC and neighbouring regions at 82 min on the genetic map of E. coli. A pulse label in vitro is incorporated into the same piece of DNA, but the center of this region, i.e. the EcoR1 fragment of 8.6 kbp length which contains the oriC region (Marsh and Worcel, 1977; v. Meyenburg et al., 1977; Yasuda and Hirota, 1977) is missing.

Replication of bacteriophage T4 DNA in vitro. I. Basic properties of the system

A new in vitro system for T4 DNA replication was developed by concentrating cell lysates on cellophane disks. The time course of [3H]dTTP incorporation into DNA by the system was separated into two phases: one was a very rapid incorporation which was terminated within 2 min (phase I reaction), and the other was a slow but continuous incorporation thereafter (phase II reaction). More than half of the phase I reaction product was Escherichia coli DNA, but the phase II reaction was mostly T4 DNA. Phase II reaction required four deoxyribonucleoside triphosphates, ATP, Mg2+, and KCl. 5-Hydroxymethyldeoxycytidine triphosphate was essential for the reaction and not substitutable by dCTP. The presence of KCN or NaN3 in the reaction mixture did not interfere with [3H]dTTP incorporation, but the addition of deoxyribonuclease completely degraded the system. Alkaline sucrose sedimentation analysis of phage II reaction product revealed that phase II reaction proceeded by the discontinuous mode of DNA replication as in vivo. After T4 infection, the activity for phase II reaction appeared in parallel with the activity of T4 phage DNA replication in vivo.

Mode of extension of the daughter strands in the replication of closed circular mitochondrial DNA in vitro

1. The properties of nascent DNA in the replicative closed circular mitochondrial DNA were examined in the in vitro system of discontinuous replication, using newborn rat liver mitochondria containing endogeneous DNA templates and enzymes. 2. The nascent DNA associated with the closed circular DNA fraction was found to be of two types; one class consisted of the fragments, and the other of the higher molecular-weight DNAs. Data from pulse and chase experiments indicate that the fragments were initially synthesized and subsequently converted into both heavy and light strands of the higher molecular-weight DNAs in an asymmetrical mode. 3. DNA - DNA hybridization experiments revealed that half of the fragments at least were copies of complementary parts of the parental DNA. 4. Based on the present in vitro data, a tentative structure of the replicating region and its expansion are discussed.

Direction of deoxyribonucleic acid replication in Escherichia coli under various conditions of cell growth

The direction of chromosome replication in a temperature-sensitive initiation mutant of Escherichia coli (CT28) is shown autoradiographically to be bidirectional. This mode of replication persists even when the rate of replication is reduced by slow growth in succinate minimal medium or in the presence of chloramphenicol. Therefore, although the rate of replication can be affected by certain physiological stimuli, the topology of replication need not be.

DNA synthesis in isolated HeLa cell nuclei. Evidence for in vitro initiation of synthesis of small pieces of DNA and their subsequent ligation

Optimum conditions for a DNA synthesizing system based on isolated nuclei have been described (Krokan, H., Bjorklid, E., and Prydz, H. (1975), Biochemistry, preceding paper in this issue) [3H]TTP-labeled nascent DNA produced during very short pulses was analyzed by centrifugation in alkaline sucrose gradients. More than 80% of the radioactivity appeared in 2-4S pieces (primary DNA pieces). It would therefore seem that the synthesis of DNA is discontinuous both in the 5' leads to 3' and in the 3' leads to 5' directions. The size of the primary DNA pieces increases from 2-4 S up to 14 S with increasing pulse length. Evidence is presented that this increase is not caused by ligation between 2-4S primary pieces. Pulse-chase experiments showed that in this nuclear system primary pieces were ligated to a product generally larger than 30 S. Evidence is also given for the initiation of primary DNA pieces in vitro.

DNA replication in Physarum polycephalum: characterization of replication products in vivo

Synchronous plasmodia of Physarum polycephalum in DNA synthesis were pulse-labelled with [oH]- thymidine for time periods of 15 seconds up to 9 minutes, or given a 30 seconds pulse followed by chase periods of 9 minutes up to 6 hours. Sedimentation analysis in alkaline sucrose gradients revealed at least five species of single stranded DNA14 molecules in the pulse experiments. Co-sedimentation of [14C]-labelled phage-DNA gave relative S-values of 5-7, 13-15, 23-25, 30 and 33-35 for these DNA molecules, all of which can be chased into DNA of higher molecular weight.

Adenovirus type 2 DNA replication. I. Evidence for discontinuous DNA synthesis

Isolated nuclei from adenovirus type 2-infected HeLa cells catalyze the incorporation of all four deoxyribonucleoside triphosphates into viral DNA. The observed DNA synthesis occurs via a transient formation of DNA fragments with a sedimentation coefficient of 10S. The fragments are precursors to unit-length viral DNA, they are self-complementary to an extent of at least 70%, and they are distributed along most of the viral chromosome. In addition, accumulation of 10S DNA fragments is observed either in intact, virus-infected HeLa cells under conditions where viral DNA synthesis is inhibited by hydroxyurea or in isolated nuclei from virus-infected HeLa cells at low concentrations of deoxyribonucleotides. Under these suboptimal conditions for DNA synthesis in isolated nuclei, ribonucleoside triphosphates determine the size distribution of DNA intermediates. The evidence presented suggests that a ribonucleoside-dependent initiation step as well at two DNA polymerase catalyzed reactions are involved in the discontinuous replication of adenovirus type 2 DNA.

In vivo repair of the single-strand interruptions contained in bacteriophage T5 DNA

Bacteriophage T5 is known to contain several unique single-strand interruptions in only one strand of the duplex DNA. Analysis of labeled parental phage DNA from infected Escherichia coli shows that these nicks are repaired in vivo to yield intact double-stranded molecules. Sealing begins at about 6 min after infection and is independent of DNA replication. Repair may be an ordered process that starts at a unique end of the molecule.

Size distribution and molecular polarity of newly replicated DNA in Escherichia coli

Newly synthesized DNA, in E. coli lysogenic for the phage lambda, was labeled by short pulses of [(3)H]-thymidine, isolated, and separated on the basis of size by alkaline sucrose density gradient centrifugation. The molecular polarity of this DNA was determined by hybridization with each of the separated strands of lambda DNA. The results show that, in the 3' to 5' direction, replication proceeds by synthesis of short chains that are subsequently joined to long DNA. This is true for both a polA(+) and a polA(-) strain. (The polA locus codes for DNA polymerase I.) In the 5' to 3' direction, replication proceeds continuously, by addition of nucleotides to long DNA, for the polA(+) strain. In the polA(-) strain, however, replication in the 5' to 3' direction is also discontinuous, but the discontinuities are 1-40 times less frequent than in the other direction.

Mechanism of DNA replication in Drosophila chromosomes: structure of replication forks and evidence for bidirectionality

The replicating chromosomal DNA in Drosophila melanogaster cleavage nuclei has been visualized in the electron microscope as a serial array of closely spaced replicated regions created by pairs of diverging replication forks. The fine structure of the forks is very similar to that observed for the replication forks of bidirectionally replicating bacteriophage DNAs. However, the mean length of the single-stranded gaps in Drosophila forks is less than 200 nucleotide residues, much shorter than the gaps in phage forks. This difference in gap length corresponds to the observed difference in the size of Okazaki fragments from Drosophila and phage.