No abstract

We earlier reported that Escherichia coli single-stranded DNA-binding protein (SSB) bound in a fixed position to the stem-loop structure of the origin of complementary DNA strand synthesis in phage G4 (G4ori c ), leaving stem-loop I and the adjacent 5 CTG 3, the primer RNA initiation site, as an SSB-free region (W. Sun and G. N. Godson Primase is required to initiate DNA replication in prokaryotes and eucaryotes. It functions to synthesize short primer RNAs (pRNA) on the separated strands of the replication fork, which provide primed templates with free 3Ј-OH terminus for DNA polymerase to elongate and synthesize the complementary DNA strand (4,13,14,25). Several prokaryotic priming systems have been established to study Escherichia coli primase (referred to as primase in this paper) function in vitro, such as cloned E. coli chromosomal origin of DNA replication (oriC) (9, 24), the origin of complementary DNA strand synthesis (ori c ) of single-stranded (ss) E. coli phages (e.g., X174[1]; G4 [3,25]; and st-1, ␣3, and K [2, 17]), double-stranded phage (e.g., [26]), and E. coli plasmid origins of DNA replication (e.g., R1 [11]). The simplest system is ss phage ori c exemplified by phage G4. In this system, primase in association with E. coli ss DNA-binding protein (SSB) synthesizes 29 nucleotides (nt) of pRNA from a unique 5Ј CTG 3Ј sequence in a region of the viral DNA strand called G4ori c (10). This is in contrast to other priming systems in which primase acts in concert with several other initiation proteins and synthesizes pRNA at many sites on the template DNA, although almost always initiating at 5Ј CTG 3Ј sequence (27). The G4ori c has been extensively studied as a model origin of DNA replication. Deletion studies have shown that the G4ori c region is a 140-nt, noncoding sequence containing potential secondary structure: stem-loops I, II, and III (5). Primase synthesizes pRNA at a 5Ј CTG 3Ј site, close to the 3Ј side of stem-loop I (7,8,18). Initiation of DNA replication in G4-related ss phages ␣3, st-1, and K (17) is similar to that of G4ori c .Several early studies provided information on the interaction of primase with the ori c DNA and SSB. A nuclease footprinting study on the interaction of primase with SSB-coated Kori c demonstrated that regions of the stem-loop structure were involved in primase binding (16). Then a stoichiometric study using gel filtration reported that two primase molecules bound on one G4ori c that was cloned into phage M13 ss-DNA (19). In a more recent study (21) to dissect the interactions of SSB and primase with G4ori c , we have demonstrated that two SSB tetramers bound on G4ori c in a fixed position. This resulted in the formation of a unique structure with two SSB tetramers (i.e., an SSB octamer) bound to the stem-loop region and other SSB tetramers bound on the 5Ј and 3Ј sides, leaving approximately 30 nt of free uncoated ss-DNA between the central SSB octamer and the adjacent SSB tetramers (or octamers). The 5Ј CTG 3Ј pRNA initiation site was located in the SSB-free linke...

Endoproteinase Asp-N (20). The number of cysteine residues (five) in primase is high for an E. coli protein and the Cys-Xaa-XaaHis-Xaal7-Cys-Xaa-Xaa-Cys motif close to the N terminus has been proposed as a zinc binding site (21). This may be a zinc finger similar to that proposed for T7 gp4 primase (22). The N terminus of primase appears to be important for its function because deletion of the first 27 N-terminal amino acids destroys the primase activity, even though the proposed zinc binding site is left intact (21).As part of a project to study the x-ray crystallographic structure ofprimase, partial proteolytic digestion was used to investigate whether primase was made up ofdiscrete physical domains. As reported below, primase contains a major physical N-terminal domain consisting of approximately the first two-thirds of the polypeptide chain, which retains the pRNA synthetic activity of intact primase in the G4ori_/SSB/ primase pRNA synthesis system. The C-terminal region of primase, however, although not required for pRNA synthesis, appears to be required for synthesis of full-length pRNA molecules, suggesting that it is required for either processivity or regulation of primase catalytic activity.MATERIALS AND METHODS Cells and Plasmids. The primase-encoding plasmid pGNG1 was constructed by inserting the wild-type E. coli dnaG gene into the T7 polymerase pET3d vector overexpression system (23) and was described in a previous paper (24). Primase was overexpressed in BL21(DE3) or the related BL21(DE3)-pLysS or pLysE strains (23). pGW398, pGW571, and pGW541, which code for C-terminal deletion mutants of primase (see below), were overexpressed in KY1421 (ilv,thr, metE,trpE9829(am),tyr(am),thy,supF-Ts6), which contains the dnaGam24 mutation (25) with a temperature-sensitive suppressor. The dnaGam24(DE3) strain was constructed by inserting the T7 polymerase gene DE3 into KY1421 using TnlO as described (23). The dnaGam24 mutation is at codon 32 of dnaG (26).Proteinase Digestion. Purified wild-type primase and the two C-terminal deletion mutants of primase, P571 and P541 Abbreviations: pRNA, primer RNA; ssDNA, single-stranded DNA; SSB, single-stranded DNA binding protein.