Mutational analysis of primosome assembly sites. Evidence for alternative DNA structures

Replication Restart in Bacteria

In bacteria, replication forks assembled at a replication origin travel to the terminus, often a few megabases away. They may encounter obstacles that trigger replisome disassembly, rendering replication restart from abandoned forks crucial for cell viability. During the past 25 years, the genes that encode replication restart proteins have been identified and genetically characterized. In parallel, the enzymes were purified and analyzed in vitro, where they can catalyze replication initiation in a sequence-independent manner from fork-like DNA structures. This work also revealed a close link between replication and homologous recombination, as replication restart from recombination intermediates is an essential step of DNA double-strand break repair in bacteria and, conversely, arrested replication forks can be acted upon by recombination proteins and converted into various recombination substrates. In this review, we summarize this intense period of research that led to the characterization of the ubiquitous replication restart protein PriA and its partners, to the definition of several replication restart pathways in vivo, and to the description of tight links between replication and homologous recombination, responsible for the importance of replication restart in the maintenance of genome stability.

The Escherichia coli primosomal DnaT protein exists in solution as a monomer-trimer equilibrium system

The oligomerization reaction of the Escherichia coli DnaT protein has been quantitatively examined using fluorescence anisotropy and analytical ultracentrifugation methods. In solution, DnaT exists as a monomer-trimer equilibrium system. At the estimated concentration in the E. coli cell, DnaT forms a mixture of the monomer and trimer states with a 3:1 molar ratio. In spite of the modest affinity, the trimerization is a highly cooperative process, without the detectable presence of the intervening dimer. The DnaT monomer consists of a large N-terminal core domain and a small C-terminal region. The removal of the C-terminal region dramatically affects the oligomerization process. The isolated N-terminal domain forms a dimer instead of the trimer. These results indicate that the DnaT monomer possesses two structurally different, interacting sites. One site is located on the N-terminal domain, and two monomers, in the trimer, are associated through their binding sites located on that domain. The C-terminal region forms the other interacting site. The third monomer is engaged through the C-terminal regions. Surprisingly, the high affinity of the N-terminal domain dimer indicates that the DnaT monomer undergoes a conformational transition upon oligomerization, involving the C-terminal region. These data and the high specificity of the trimerization reaction, i.e., lack of any oligomers higher than a trimer, indicate that each monomer in the trimer is in contact with the two remaining monomers. A model of the global structure of the DnaT trimer based on the thermodynamic and hydrodynamic data is discussed.

The N-terminal domain of the Escherichia coli PriA helicase contains both the DNA- and nucleotide-binding sites. Energetics of domain--DNA interactions and allosteric effect of the nucleotide cofactors

Functional interactions of the Escherichia coli PriA helicase 181N-terminal domain with the DNA and nucleotide cofactors have been quantitatively examined. The isolated 181N-terminal domain forms a stable dimer in solution, most probably reflecting the involvement of the domain in specific cooperative interactions of the intact PriA protein--double-stranded DNA (dsDNA) complex. Only one monomer of the domain dimer binds the DNA; i.e., the dimer has one effective DNA-binding site. Although the total site size of the dimer--single-stranded DNA (ssDNA) complex is ~13 nucleotides, the DNA-binding subsite engages in direct interactions with approximately five nucleotides. A small number of interacting nucleotides indicates that the DNA-binding subsites of the PriA helicase, i.e., the strong subsite on the helicase domain and the weak subsite on the N-terminal domain, are spatially separated in the intact enzyme. Contrary to current views, the subsite has an only slight preference for the 3'-end OH group of the ssDNA and lacks any significant base specificity, although it has a significant dsDNA affinity. Unlike the intact helicase, the DNA-binding subsite of the isolated domain is in an open conformation, indicating the presence of the direct helicase domain--N-terminal domain interactions. The discovery that the 181N-terminal domain possesses a nucleotide-binding site places the allosteric, weak nucleotide-binding site of the intact PriA on the N-terminal domain. The specific effect of ADP on the domain DNA-binding subsite indicates that in the intact helicase, the bound ADP not only opens the DNA-binding subsite but also increases its intrinsic DNA affinity.

Binding of two PriA-PriB complexes to the primosome assembly site initiates primosome formation

A direct quantitative analysis of the initial steps in primosome assembly, involving PriA and PriB proteins and the minimal primosome assembly site (PAS) of phage ϕX174, has been performed using fluorescence intensity, fluorescence anisotropy titration, and fluorescence resonance energy transfer techniques. We show that two PriA molecules bind to the PAS at both strong and weak binding sites on the DNA, respectively, without detectable cooperative interactions. Binding of the PriB dimer to the PriA-PAS complex dramatically increases PriA's affinity for the strong site, but only slightly affects its affinity for the weak site. Associations with the strong and weak sites are driven by apparent entropy changes, with binding to the strong site accompanied by a large unfavorable enthalpy change. The PriA-PriB complex, formed independently of the DNA, is able to directly recognize the PAS without the preceding the binding of PriA to the PAS. Thus, the high-affinity state of PriA for PAS is generated through PriA-PriB interactions. The effect of PriB is specific for PriA-PAS association, but not for PriA-double-stranded DNA or PriA-single-stranded DNA interactions. Only complexes containing two PriA molecules can generate a profound change in the PAS structure in the presence of ATP. The obtained results provide a quantitative framework for the elucidation of further steps in primosome assembly and for quantitative analyses of other molecular machines of cellular metabolism.

PriA: at the crossroads of DNA replication and recombination

PriA is a single-stranded DNA-dependent ATPase, DNA translocase, and DNA helicase that was discovered originally because of its requirement in vitro for the conversion of bacteriophage phi X174 viral DNA to the duplex replicative form. Studies demonstrated that PriA catalyzes the assembly of a primosome, a multiprotein complex that primes DNA synthesis, on phi X174 DNA. The primosome was shown to be capable of providing both the DNA unwinding function and the Okazaki fragment priming function required for replication fork progression. However, whereas seven proteins, PriA, PriB, PriC, DnaT, DnaB, DnaC, and DnaG, were required for primosome assembly on phi X174 DNA, only DnaB, DnaC, and DnaG were required for replication from oriC, suggesting that the other proteins were not involved in chromosomal replication. Strains carrying priA null mutations, however, were constitutively induced for the SOS response, and were defective in homologous recombination, repair of UV-damaged DNA, and double-strand breaks, and both induced and constitutive stable DNA replication. The basis for this phenotype can now be explained by the ability of PriA to load replication forks at a D loop, an intermediate that forms during homologous recombination, double-strand break-repair, and stable DNA replication. Thus, a long-theorized connection between recombination and replication is demonstrated.

Two modes of PriA binding to DNA

The role of PriA, required for the assembly of the phiX174-type primosome on DNA, in cellular DNA replication has been unclear since its discovery. Recent evidence, based on the phenotypes of strains carrying priA null mutations, has led to proposals that the primosome assembly activity of PriA was required to load replication forks at intermediates such as D loops during homologous recombination. McGlynn et al. (McGlynn, P., Al-Deib, A. A., Liu, J., Marians, K. J., and Lloyd, R. G. (1997) J. Mol. Biol. 270, 212-221) demonstrated that PriA could, in fact, bind D loops. We show here that there are two modes of stable binding of PriA to DNA. One mode, in which the enzyme binds 3'-single-stranded extensions from duplex DNAs, presumably reflects the 3' --> 5' DNA helicase activity of PriA. The D loop DNA binding activity of PriA can be accounted for by the second mode, where the enzyme binds bent DNA at three strand junctions.

Structure of the Escherichia coli primase/single-strand DNA-binding protein/phage G4oric complex required for primer RNA synthesis

Escherichia coli primase/SSB/single-stranded phage G4oric is a simple system to study how primase interacts with DNA template to synthesize primer RNA for initiation of DNA replication. By a strategy of deletion analysis and antisense oligonucleotide protection on small single-stranded G4oric fragments, we have identified the DNA sequences required for binding primase and the critical location of single-strand DNA-binding (SSB) protein. Together with the previous data, we have defined the structure of the primase/SSB/G4oric priming complex. Two SSB tetramers bind to the G4oric secondary structure, which dictates the spacing of 3' and 5' bound adjacent SSB tetramers and leaves SSB-free regions on both sides of the stem-loop structure. Two primase molecules then bind separately to specific DNA sequences in the 3' and 5' SSB-free G4oric regions. Binding of the 3' SSB tetramer, upstream of the primer RNA initiation site, is also necessary for priming. The generation of a primase-recognition target by SSB phasing at DNA hairpin structures may be applicable to the binding of initiator proteins in other single-stranded DNA priming systems. Novel techniques used in this study include antisense oligonucleotide protection and RNA synthesis on an SSB-melted, double-stranded DNA template.

A DNA structure is required for geminivirus replication origin function

The genome of the geminivirus tomato golden mosaic virus (TGMV) consists of two single-stranded circular DNAs, A and B, that replicate through a rolling-circle mechanism in nuclei of infected plant cells. The TGMV origin of replication is located in a conserved 5' intergenic region and includes at least two functional elements: the origin recognition site of the essential viral replication protein, AL1, and a sequence motif with the potential to form a hairpin or cruciform structure. To address the role of the hairpin motif during TGMV replication, we constructed a series of B-component mutants that resolved sequence changes from structural alterations of the motif. Only those mutant B DNAs that retained the capacity to form the hairpin structure replicated to wild-type levels in tobacco protoplasts when the viral replication proteins were provided in trans from a plant expression cassette. In contrast, the same B DNAs replicated to significantly lower levels in transient assays that included replicating, wild-type TGMV A DNA. These data established that the hairpin structure is essential for TGMV replication, whereas its sequence affects the efficiency of replication. We also showed that TGMV AL1 functions as a site-specific endonuclease in vitro and mapped the cleavage site to the loop of the hairpin. In vitro cleavage analysis of two TGMV B mutants with different replication phenotypes indicated that there is a correlation between the two assays for origin activity. These results suggest that the in vivo replication results may reflect structural and sequence requirements for DNA cleavage during initiation of rolling-circle replication.

Comparative analysis of functional and structural features in the primase-dependent priming signals, G sites, from phages and plasmids

The primase-dependent priming signals, G sites, are directly recognized by the Escherichia coli primase (dnaG gene product) and conduct the synthesis of primer RNAs. In nucleotide sequence and secondary structure, there is no striking resemblance between the phage- and plasmid-derived G sites, except for the limited sequence homology near the start position of primer RNA synthesis. In this study, we analyzed the structure and function of a G site of plasmid R100, G site (R100), and discovered the necessity of the coexistence of two domains (domains I and III), which contains blocks A, B, and C, which are nucleotide sequences highly conserved among the plasmid-derived G sites. However, neither the internal region, domain II, between domains I and III nor the potential secondary structure proposed by Bahk et al. (J. D. Bahk, N. Kioka, H. Sakai, and T. Komano, Plasmid 20:266-270, 1988) is essential for single-stranded DNA initiation activity. Furthermore, chimeric G sites constructed between a G site of phage G4, G site(G4), and G site(R100) maintained significant single-stranded DNA initiation activities. These results strongly suggest that phage- and plasmid-derived G sites have functionally equivalent domains. The primase-dependent priming mechanisms of phage- and plasmid-derived G sites are discussed.

Mutational analysis of the specific priming signal essential for DNA replication of the broad host-range plasmid RSF1010

To analyze the RSF1010-specific priming mechanism, a library of randomly mutagenized ssiA sequences was constructed by chemical synthesis using mixed nucleotide phosphoramidites. Synthetic ssiA sequences with the single base-substitutions were assayed for the SSI activity in E. coli JM109 expressing RepB' primase. It was demonstrated that the activity of ssiA was damaged markedly by single base-substitutions within the possible stem-loop structure and its 3'-flanking region. It is conceivable that these domains are critical in recognition and primer synthesis by RepB' primase.

The single-stranded DNA-binding protein of Escherichia coli

The single-stranded DNA-binding protein (SSB) of Escherichia coli is involved in all aspects of DNA metabolism: replication, repair, and recombination. In solution, the protein exists as a homotetramer of 18,843-kilodalton subunits. As it binds tightly and cooperatively to single-stranded DNA, it has become a prototypic model protein for studying protein-nucleic acid interactions. The sequences of the gene and protein are known, and the functional domains of subunit interaction, DNA binding, and protein-protein interactions have been probed by structure-function analyses of various mutations. The ssb gene has three promoters, one of which is inducible because it lies only two nucleotides from the LexA-binding site of the adjacent uvrA gene. Induction of the SOS response, however, does not lead to significant increases in SSB levels. The binding protein has several functions in DNA replication, including enhancement of helix destabilization by DNA helicases, prevention of reannealing of the single strands and protection from nuclease digestion, organization and stabilization of replication origins, primosome assembly, priming specificity, enhancement of replication fidelity, enhancement of polymerase processivity, and promotion of polymerase binding to the template. E. coli SSB is required for methyl-directed mismatch repair, induction of the SOS response, and recombinational repair. During recombination, SSB interacts with the RecBCD enzyme to find Chi sites, promotes binding of RecA protein, and promotes strand uptake.

Distinct functional contributions of three potential secondary structures in the phage G4 origin of complementary DNA strand synthesis

Three potential secondary structures, stem-loops I, II, and III, are contained in the phage G4 origin of complementary DNA strand synthesis, G4oric, and are believed to be involved in its recognition by dnaG-encoded primase and the synthesis of primer RNA. In a previous publication [Sakai et al., Gene 71 (1988) 323-330], we suggested that base pairing between the loops of stem-loops I, and II, and/or II and III, might play a role in G4oric function. To test this hypothesis, site-directed mutagenesis was used to construct mutants which carried base substitutions in loops I, II and III that destroyed possible interloop base pairing. These mutations, however, did not seriously affect G4oric activity. This indicates that base pairing between the loops is not essential for G4oric functional activity, and also that base substitutions which do not affect the secondary structure of stem-loops I, II and III, do not affect G4oric activity. To complete an analysis of the effects of altering the structure of the G4oric stem-loops, insertions were made into stem-loop III. In contrast to stem-loops I and II, all insertions into stem-loop III destroyed in vivo G4oric activity.