Denaturation of the simian virus 40 origin of replication mediated by human replication protein A.

The initiation of simian virus 40 (SV40) replication requires recognition of the viral origin of replication (ori) by SV40 T antigen, followed by denaturation of ori in a reaction dependent upon human replication protein A (hRPA). To understand how origin denaturation is achieved, we constructed a 48-bp SV40 "pseudo-origin" with a central 8-nucleotide (nt) bubble flanked by viral sequences, mimicking a DNA structure found within the SV40 T antigen-ori complex. hRPA bound the pseudo-origin with similar stoichiometry and an approximately fivefold reduced affinity compared to the binding of a 48-nt single-stranded DNA molecule. The presence of hRPA not only distorted the duplex DNA flanking the bubble but also resulted in denaturation of the pseudo-origin substrate in an ATP-independent reaction. Pseudo-origin denaturation occurred in 7 mM MgCl2, distinguishing this reaction from Mg2+-independent DNA-unwinding activities previously reported for hRPA. Tests of other single-stranded DNA-binding proteins (SSBs) revealed that pseudo-origin binding correlates with the known ability of these SSBs to support the T-antigen-dependent origin unwinding activity. Our results suggest that hRPA binding to the T antigen-ori complex induces the denaturation of ori including T-antigen recognition sequences, thus releasing T antigen from ori to unwind the viral DNA. The denaturation activity of hRPA has the potential to play a significant role in other aspects of DNA metabolism, including DNA repair.     

Unwinding of origin-specific structures by human replication protein A occurs in a two-step process.

The simian virus 40 (SV40) large tumor antigen(T antigen) has been shown to induce the melting of 8 bp within the SV40 origin of replication. We found previously that a 'pseudo-origin' DNA molecule (PO-8) containing a central 8 nt single-stranded DNA (ssDNA) bubble was efficiently bound and denatured by human replication protein A (hRPA). To understand the mechanism by which hRPA denatures these pseudo-origin molecules, as well as the role that hRPA plays during the initiation of SV40 DNA replication, we characterized the key parameters for the pseudo-origin binding and denaturation reactions. The dissociation constant of hRPA binding to PO-8 was observed to be 7.7 x 10(-7) M, compared to 9.0 x 10(-8) M for binding to an identical length ssDNA under the same reaction conditions. The binding and denaturation of PO-8 occurred with different kinetics with the rate of binding determined to be approximately 4-fold greater than the rate of denaturation. Although hRPA binding to PO-8 was relatively temperature independent, an increase in incubation temperature from 4 to 37 degreesC stimulated denaturation nearly 4-fold. At 37 degreesC, denaturation occurred on approximately 1/3 of those substrate molecules bound by hRPA, showing that hRPA can bind the pseudo-origin substrate without causing its complete denaturation. Tests of other single-stranded DNA-binding proteins (SSBs) over a range of SSB concentrations revealed that the ability of the SSBs to bind the pseudo-origin substrate, rather than denature the substrate, correlated best with the known ability of these SSBs to support the T antigen-dependent SV40 origin-unwinding activity. Our data indicate that hRPA first binds the DNA substrate using a combination of contacts with the ssDNA bubble and duplex DNA flanks and then, on only a fraction of the bound substrate molecules, denatures the DNA substrate.     

Single-stranded-DNA binding alters human replication protein A structure and facilitates interaction with DNA-dependent protein kinase.

Human replication protein A (hRPA) is an essential single-stranded-DNA-binding protein that stimulates the activities of multiple DNA replication and repair proteins through physical interaction. To understand DNA binding and its role in hRPA heterologous interaction, we examined the physical structure of hRPA complexes with single-stranded DNA (ssDNA) by scanning transmission electron microscopy. Recent biochemical studies have shown that hRPA combines with ssDNA in at least two binding modes: by interacting with 8 to 10 nucleotides (hRPA8nt) and with 30 nucleotides (hRPA30nt). We find the relatively unstable hRPA8nt complex to be notably compact with many contacts between hRPA molecules. In contrast, on similar lengths of ssDNA, hRPA30nt complexes align along the DNA and make few intermolecular contacts. Surprisingly, the elongated hRPA30nt complex exists in either a contracted or an extended form that depends on ssDNA length. Therefore, homologous-protein interaction and available ssDNA length both contribute to the physical changes that occur in hRPA when it binds ssDNA. We used activated DNA-dependent protein kinase as a biochemical probe to detect alterations in conformation and demonstrated that formation of the extended hRPA30nt complex correlates with increased phosphorylation of the hRPA 29-kDa subunit. Our results indicate that hRPA binds ssDNA in a multistep pathway, inducing new hRPA alignments and conformations that can modulate the functional interaction of other factors with hRPA.     

Model for T-antigen-dependent melting of the simian virus 40 core origin based on studies of the interaction of the beta-hairpin with DNA

The interaction of simian virus 40 (SV40) T antigen (T-ag) with the viral origin has served as a model for studies of site-specific recognition of a eukaryotic replication origin and the mechanism of DNA unwinding. These studies have revealed that a motif termed the "beta-hairpin" is necessary for assembly of T-ag on the SV40 origin. Herein it is demonstrated that residues at the tip of the "beta-hairpin" are needed to melt the origin-flanking regions and that the T-ag helicase domain selectively assembles around one of the newly generated single strands in a manner that accounts for its 3'-to-5' helicase activity. Furthermore, T-ags mutated at the tip of the "beta-hairpin" are defective for oligomerization on duplex DNA; however, they can assemble on hybrid duplex DNA or single-stranded DNA (ssDNA) substrates provided the strand containing the 3' extension is present. Collectively, these experiments indicate that residues at the tip of the beta-hairpin generate ssDNA in the core origin and that the ssDNA is essential for subsequent oligomerization events.     

Synthetic DNA Replication Bubbles Bound and Unwound with Twofold Symmetry by a Simian Virus 40 T-Antigen Double Hexamer

Dimerization of simian virus 40 T-antigen hexamers (TAg_H) into double hexamers (TAg_DH) on model DNA replication forks has been found to greatly stimulate T-antigen DNA helicase activity. To explore the interaction of TAg_DH with DNA during unwinding, we examined the binding of TAg_DH to synthetic DNA replication bubbles. Tests of replication bubble substrates containing different single-stranded DNA (ssDNA) lengths indicated that efficient formation of a TAg_DH requires ≥40 nucleotides (nt) of ssDNA. DNase I probing of a substrate containing a 60-nt ssDNA bubble complexed with a TAg_DH revealed that T antigen bound the substrate with twofold symmetry. The strongest protection was observed over the 5′ junction on each strand, with 5 bp of duplex DNA and ~17 nt of adjacent ssDNA protected from nuclease cleavage. Stimulation of the T-antigen DNA helicase activity by an increase in ATP concentration caused the protection to extend in the 5′ direction into the duplex region, while resulting in no significant changes to the 3′ edge of strongest protection. Our data indicate that each TAg_H encircles one ssDNA strand, with a different strand bound at each junction. The process of DNA unwinding results in each TAg_H interacting with a greater length of DNA than was initially bound, suggesting the generation of a more highly processive helicase complex.     

Human Rad52 binds and wraps single-stranded DNA and mediates annealing via two hRad52–ssDNA complexes

Rad52 promotes the annealing of complementary strands of DNA bound by replication protein A (RPA) during discrete repair pathways. Here, we used a fluorescence resonance energy transfer (FRET) between two fluorescent dyes incorporated into DNA substrates to probe the mechanism by which human Rad52 (hRad52) interacts with and mediates annealing of ssDNA–hRPA complexes. Human Rad52 bound ssDNA or ssDNA–hRPA complex in two, concentration-dependent modes. At low hRad52 concentrations, ssDNA was wrapped around the circumference of the protein ring, while at higher protein concentrations, ssDNA was stretched between multiple hRad52 rings. Annealing by hRad52 occurred most efficiently when each complementary DNA strand or each ssDNA–hRPA complex was bound by hRad52 in a wrapped configuration, suggesting homology search and annealing occur via two hRad52–ssDNA complexes. In contrast to the wild type protein, hRad52^RQK/AAA and hRad52^1–212 mutants with impaired ability to bind hRPA protein competed with hRPA for binding to ssDNA and failed to counteract hRPA-mediated duplex destabilization highlighting the importance of hRad52-hRPA interactions in promoting efficient DNA annealing.     

Insights into hRPA32 C-terminal domain--mediated assembly of the simian virus 40 replisome

Simian virus 40 (SV40) provides a model system for the study of eukaryotic DNA replication, in which the viral protein, large T antigen (Tag), marshals human proteins to replicate the viral minichromosome. SV40 replication requires interaction of Tag with the host single-stranded DNA-binding protein, replication protein A (hRPA). The C-terminal domain of the hRPA32 subunit (RPA32C) facilitates initiation of replication, but whether it interacts with Tag is not known. Affinity chromatography and NMR revealed physical interaction between hRPA32C and the Tag origin DNA-binding domain, and a structural model of the complex was determined. Point mutations were then designed to reverse charges in the binding sites, resulting in substantially reduced binding affinity. Corresponding mutations introduced into intact hRPA impaired initiation of replication and primosome activity, implying that this interaction has a critical role in assembly and progression of the SV40 replisome.     

Diffusion of Human Replication Protein A along Single-Stranded DNA

Replication protein A (RPA) is a eukaryotic single-stranded DNA (ssDNA) binding protein that plays critical roles in most aspects of genome maintenance, including replication, recombination and repair. RPA binds ssDNA with high affinity, destabilizes DNA secondary structure and facilitates binding of other proteins to ssDNA. However, RPA must be removed from or redistributed along ssDNA during these processes. To probe the dynamics of RPA–DNA interactions, we combined ensemble and single-molecule fluorescence approaches to examine human RPA (hRPA) diffusion along ssDNA and find that an hRPA heterotrimer can diffuse rapidly along ssDNA. Diffusion of hRPA is functional in that it provides the mechanism by which hRPA can transiently disrupt DNA hairpins by diffusing in from ssDNA regions adjacent to the DNA hairpin. hRPA diffusion was also monitored by the fluctuations in fluorescence intensity of a Cy3 fluorophore attached to the end of ssDNA. Using a novel method to calibrate the Cy3 fluorescence intensity as a function of hRPA position on the ssDNA, we estimate a one-dimensional diffusion coefficient of hRPA on ssDNA of D₁ ~ 5000 nt² s^− 1 at 37 °C. Diffusion of hRPA while bound to ssDNA enables it to be readily repositioned to allow other proteins access to ssDNA.     

Characterization of the DNA-unwinding activity of human RECQ1, a helicase specifically stimulated by human replication protein A

The RecQ helicases are involved in several aspects of DNA metabolism. Five members of the RecQ family have been found in humans, but only two of them have been carefully characterized, BLM and WRN. In this work, we describe the enzymatic characterization of RECQ1. The helicase has 3' to 5' polarity, cannot start the unwinding from a blunt-ended terminus, and needs a 3'-single-stranded DNA tail longer than 10 nucleotides to open the substrate. However, it was also able to unwind a blunt-ended duplex DNA with a "bubble" of 25 nucleotides in the middle, as previously observed for WRN and BLM. We show that only short DNA duplexes (<30 bp) can be unwound by RECQ1 alone, but the addition of human replication protein A (hRPA) increases the processivity of the enzyme (>100 bp). Our studies done with Escherichia coli single-strand binding protein (SSB) indicate that the helicase activity of RECQ1 is specifically stimulated by hRPA. This finding suggests that RECQ1 and hRPA may interact also in vivo and function together in DNA metabolism. Comparison of the present results with previous studies on WRN and BLM provides novel insight into the role of the N- and C-terminal domains of these helicases in determining their substrate specificity and in their interaction with hRPA.     

Strand and face: the topography of interactions between the SV40 origin of replication and T-antigen during the initiation of replication.

The mechanism by which a replicator (origin of replication) becomes denatured during the initiation of replication is not understood for any prokaryotic or eukaryotic system. To address this question, we chemically probed the molecular contacts on the SV40 origin of replication (ori) that are used by the SV40 large T-antigen and a single-stranded DNA-binding protein (SSB) during ori denaturation. Prior to the actual denaturation step, the T-antigen double hexamer bound ori utilizing sugar-phosphate contacts that were located on opposite strands in each flanking domain of ori. Each set of flanking phosphate contacts were also located on approximately opposite faces of the ori duplex. While the phosphate contacts had a 2-fold symmetry with respect to the ori center, T-antigen contacts with nucleotide bases were polar with critical interactions detected in only one of the two flanking domains. During origin denaturation catalyzed by T-antigen and a SSB, numerous new contacts to flanking phosphates were observed on the strand not initially bound by T-antigen, suggesting movement of each T-antigen hexamer outward from ori. These data suggest that T-antigen initially binds ori in a manner that facilitates transfer of each T-antigen hexamer to opposite strands during the initiation of SV40 replication.     

Solution structure of the DNA-binding domain of RPA from Saccharomyces cerevisiae and its interaction with single-stranded DNA and SV40 T antigen

Replication protein A (RPA) is a three-subunit complex with multiple roles in DNA metabolism. DNA-binding domain A in the large subunit of human RPA (hRPA70A) binds to single-stranded DNA (ssDNA) and is responsible for the species-specific RPA–T antigen (T-ag) interaction required for Simian virus 40 replication. Although Saccharomyces cerevisiae RPA70A (scRPA70A) shares high sequence homology with hRPA70A, the two are not functionally equivalent. To elucidate the similarities and differences between these two homologous proteins, we determined the solution structure of scRPA70A, which closely resembled the structure of hRPA70A. The structure of ssDNA-bound scRPA70A, as simulated by residual dipolar coupling-based homology modeling, suggested that the positioning of the ssDNA is the same for scRPA70A and hRPA70A, although the conformational changes that occur in the two proteins upon ssDNA binding are not identical. NMR titrations of hRPA70A with T-ag showed that the T-ag binding surface is separate from the ssDNA-binding region and is more neutral than the corresponding part of scRPA70A. These differences might account for the species-specific nature of the hRPA70A–T-ag interaction. Our results provide insight into how these two homologous RPA proteins can exhibit functional differences, but still both retain their ability to bind ssDNA.     

Single-stranded DNA binding proteins unwind the newly synthesized double-stranded DNA of model miniforks

Single-stranded DNA binding (SSB) proteins are essential proteins of DNA metabolism. We characterized the binding of the bacteriophage T4 SSB, Escherichia coli SSB, human replication protein A (hRPA), and human hSSB1 proteins onto model miniforks and double-stranded-single-stranded (ds-ss) junctions exposing 3' or 5' ssDNA overhangs. T4 SSB proteins, E. coli SSB proteins, and hRPA have a different binding preference for the ss tail exposed on model miniforks and ds-ss junctions. The T4 SSB protein preferentially binds substrates with 5' ss tails, whereas the E. coli SSB protein and hRPA show a preference for substrates with 3' ss overhangs. When interacting with ds-ss junctions or miniforks, the T4 SSB protein, E. coli SSB protein, and hRPA can destabilize not only the ds part of a ds-ss junction but also the daughter ds arm of a minifork. The T4 SSB protein displays these unwinding activities in a polar manner. Taken together, our results position the SSB protein as a potential key player in the reversal of a stalled replication fork and in gap repair-mediated repetitive sequence expansion.     

A Dimer of Escherichia coli UvrD is the active form of the helicase in vitro

The Escherichia coli UvrD protein is a 3' to 5' SF1 DNA helicase involved in methyl-directed mismatch repair and nucleotide excision repair of DNA. We have characterized in vitro UvrD-catalyzed unwinding of a series of 18 bp duplex DNA substrates with 3' single-stranded DNA (ssDNA) tails ranging in length from two to 40 nt. Single turnover DNA-unwinding experiments were performed using chemical quenched flow methods, as a function of both [UvrD] and [DNA] under conditions such that UvrD-DNA binding is stoichiometric. Although a single UvrD monomer binds tightly to the single-stranded/double-stranded DNA (dsDNA) junction if the 3' ssDNA tail is at least four nt, no unwinding was observed for DNA substrates with tail-lengths /=12 nt, and the unwinding amplitude displays a sigmoidal dependence on [UvrD(tot)]/[DNA(tot)]. Quantitative analysis of these data indicates that a single UvrD monomer bound at the ssDNA/dsDNA junction of any DNA substrate, independent of 3' ssDNA tail length, is not competent to fully unwind even a short 18 bp duplex DNA, and that two UvrD monomers must bind the DNA substrate in order to form a complex that is able to unwind short DNA substrates in vitro. Other proteins, including a mutant UvrD with no ATPase activity as well as a monomer of the structurally homologous E.coli Rep helicase, cannot substitute for the second UvrD monomer, suggesting a specific interaction between two UvrD monomers and that both must be able to hydrolyze ATP. Initiation of DNA unwinding in vitro appears to require a dimeric UvrD complex in which one subunit is bound to the ssDNA/dsDNA junction, while the second subunit is bound to the 3' ssDNA tail.     

DNA binding properties of the adenovirus DNA replication priming protein pTP

The precursor terminal protein pTP is the primer for the initiation of adenovirus (Ad) DNA replication and forms a heterodimer with Ad DNA polymerase (pol). Pol can couple dCTP to pTP directed by the fourth nucleotide of the viral genome template strand in the absence of other replication proteins, which suggests that pTP/pol binding destabilizes the origin or stabilizes an unwound state. We analyzed the contribution of pTP to pTP/pol origin binding using various DNA oligonucleotides. We show that two pTP molecules bind cooperatively to short DNA duplexes, while longer DNA fragments are bound by single pTP molecules as well. Cooperative binding to short duplexes is DNA sequence independent and most likely mediated by protein/protein contacts. Furthermore, we observed that pTP binds single-stranded (ss)DNA with a minimal length of approximately 35 nt and that random ssDNA competed 25-fold more efficiently than random duplex DNA for origin binding by pTP. Remarkably, short DNA fragments with two opposing single strands supported monomeric pTP binding. pTP did not stimulate, but inhibited strand displacement by the Ad DNA binding and unwinding protein DBP. These observations suggest a mechanism in which the ssDNA affinity of pTP stabilizes Ad pol on partially unwound origin DNA.     

Recognition of nucleic acid double helices by homopyrimidine 2', 5'-linked RNA.

We have studied the effect of a 2',5'-RNA third strand backbone on the stability of triple helices with a 'pyrimidine motif' targeting the polypurine strand of duplex DNA, duplex RNA and DNA/RNA hybrids. Comparative experiments were run in parallel with DNA and the regioisomeric RNA as third strands adopting the experimental design of Roberts and Crothers. The results reveal that 2',5'-RNA is indeed able to recognize double helical DNA (DD) and DNA (purine):RNA (pyrimidine) hybrids (DR). However, when the duplex purine strand is RNA and the duplex pyrimidine strand is DNA or RNA (i.e. RD or RR), triplex formation is not observed. These results exactly parallel what is observed for DNA third strands. Based on T m data, the affinities of 2',5'-RNA and DNA third strands towards DD and DR duplexes were similar. The RNA third strand formed triplexes with all four hairpins, as previously demonstrated. In analogy to the arabinose and 2'-deoxyribose third strands, the possible C2'- endo pucker of 2',5'-linked riboses together with the lack of an alpha-2'-OH group are believed to be responsible for the selective binding of 2',5'-RNA to DD and DR duplexes, over RR and RD duplexes. These studies indicate that the use of other oligonucleotide analogues will prove extremely useful in dissecting the contributions of backbone and/or sugar puckering to the recognition of nucleic acid duplexes.     

Origin auxiliary sequences can facilitate initiation of simian virus 40 DNA replication in vitro as they do in vivo.

Initiation of simian virus 40 (SV40) DNA replication is facilitated by two auxiliary sequences that flank the minimally required origin (ori) core sequence. In monkey cells, the replication rate of each of the four ori configurations changed with time after transfection in a characteristic pattern. This pattern was reproduced in an extract from SV40-infected monkey cells by varying the ratio of DNA substrate to cell extract; DNA replication in vitro depended on ori auxiliary sequences to the same extent as they did in vivo. Facilitation by ori auxiliary sequences was lost at high ratios of DNA to cell extract, revealing that the activity of these sequences required either multiple initiation factors or a molar excess of one initiation factor bound to ori. This parameter, together with ionic strength and the method used to measure DNA replication, determined the level of facilitation by ori auxiliary sequences in vitro. The activity of ori auxiliary sequences was not diminished in vivo or in vitro by increasing amounts of large tumor antigen. Therefore, ori auxiliary sequences promoted initiation of replication at some step after tumor antigen binding to ori. Furthermore, although cellular factors could modulate the activity of ori auxiliary sequences in vitro, these factors did not appear to involve nucleosome assembly because no correlation was observed between the number of nucleosomes assembled per DNA molecule and facilitation by ori auxiliary sequences. These results demonstrate that SV40 ori auxiliary sequences can function in vitro as they do in vivo and begin to elucidate their role in initiating DNA replication.     

Hepatitis C virus NS3 and simian virus 40 T antigen helicases displace streptavidin from 5'-biotinylated oligonucleotides but not from 3'-biotinylated oligonucleotides: evidence for directional bias in translocation on single-stranded DNA

Helicases are enzymes that use energy from nucleoside triphosphate hydrolysis to unwind double-stranded (ds) DNA, a process vital to virtually every phase of DNA metabolism. Helicases have been classified as either 5'-to-3' or 3'-to-5' on the basis of their ability to unwind duplex DNA adjacent to either a 5' or 3' single-stranded (ss) DNA overhang. However, there has been debate as to whether this substrate preference is indicative of unidirectional translocation on ssDNA. We developed an assay that monitors the ability of a helicase to displace streptavidin from biotinylated oligonucleotides [Morris, P. D., and Raney, K. D. (1999) Biochemistry 38, 5164-5171]. Two helicases identified as having 5'-to-3' polarity displaced streptavidin from the 3'-end of biotinylated oligonucleotides but not from the 5'-end. We performed similar experiments using the 3'-to-5' helicases from the hepatitis C virus (NS3) and SV40 virus (SV40 T antigen). NS3 and SV40 T antigen were able to displace streptavidin from a 5'-biotinylated oligonucleotide but not from a 3'-biotinylated oligonucleotide. NS3 and SV40 T antigen enhanced the spontaneous rate of dissociation of streptavidin from biotin 340-fold and 1700-fold, respectively. The ssDNA binding protein, gp32, did not enhance dissociation of streptavidin from either end of an oligonucleotide. For NS3, the rate of displacement was faster from a 5'-biotinylated 60mer than from a 5'-biotinylated 30mer. The strong directional bias in streptavidin displacement activity exhibited by each helicase is consistent with a directional bias in translocation on ssDNA. The dependence of the reaction with NS3 on the oligonucleotide length suggests that multiple NS3 monomers are necessary for optimal activity.     

DNA helicase and duplex DNA fragment unwinding activities of polyoma and simian virus 40 large T antigen display similarities and differences.

We have characterized the biochemical activities of purified polyoma (Py) large T antigen (T Ag) that was capable of mediating the replication of a plasmid containing the Py origin (ori(+) DNA) in mouse cell extracts. We report here that like the T Ag encoded by simian virus 40 (SV40), Py T Ag has DNA helicase and double-stranded DNA fragment unwinding activities. Py T Ag displaced DNA fragments greater than 1,600 nucleotides which were annealed to complementary sequences in single-stranded M13 by translocating in the 3' to 5' direction. Both helicase and double-stranded DNA fragment unwinding reactions were completely dependent upon NTP hydrolysis, displaying a strong preference for ATP and dATP. At low T Ag concentrations, significantly more Py ori(+) DNA fragment was unwound compared with a fragment lacking the replication origin. However, at higher ratios of Py T Ag to DNA, equivalent to those used in replication reactions, unwinding of both ori-containing and -lacking fragments was equally efficient. This is in contrast to SV40 T Ag which exhibited a more stringent requirement for SV40 origin sequences under similar conditions. Furthermore, some of the nucleotides that supported the helicase and unwinding activities of Py T Ag were different from those for the same SV40 T Ag reactions. We have also observed that in contrast to the very poor replication of linear SV40 ori(+) DNA by SV40 T Ag in human cell extracts, linear Py ori(+) DNA was replicated efficiently in mouse cell extracts by Py T Ag. However, despite the fact that linear Py ori(+), SV40 ori(+), and ori(-) DNA fragments could be unwound with comparable efficiency by Py T Ag, only fragments containing the Py replication origin were replicated in vitro. These results suggest that the initiation of DNA synthesis at the Py origin of replication requires features in addition to unwinding of the template.     

Initiation of SV40 DNA replication in vivo: location and structure of 5' ends of DNA synthesized in the ori region

Initiation sites for DNA synthesis were located at the resolution of single nucleotides in and about the genetically defined origin of replication (ori) in replicating SV40 DNA purified from virus-infected cells. About 50% of the DNA chains contained an oligoribonucleotide of six to nine residues covalently attached to their 5' ends. Although the RNA-DNA linkage varied, the putative RNA primer began predominantly with rA. The data reveal that initiation of DNA synthesis is promoted at a number of DNA sequences that are asymmetrically arranged with respect to ori: 5' ends of nascent DNA are located at several sites within ori, but only on the strand that also serves as the template for early mRNA, while 5' ends of nascent DNA with the opposite orientation are located only outside ori on its early gene side. This clear transition between discontinuous (initiation sites) and continuous (no initiation sites) DNA synthesis defines the origin of bidirectional replication at nucleotides 5210--5211 and demonstrates that discontinuous synthesis occurs predominantly on the retrograde arms of replication forks. Furthermore, it appears that the first nascent DNA chain is initiated within ori by the same mechanism used to initiate nascent DNA ("Okazaki fragments") throughout the genome.