Introduction

Chromosome replication is strictly regulated to ensure successful duplication of the genetic material (Costa & Diffley, 2022; Grimwade & Leonard, 2021; Kasho et al., 2023). In Escherichia coli, the ATP-bound form of the replication initiator protein DnaA (ATP-DnaA) forms a higher-order complex with DNA-bending protein IHF at the unique chromosomal origin oriC. This complex promotes local unwinding of oriC and recruits a pair of DnaB helicases, which are successively loaded to the single-stranded (ss) DNA regions in a bidirectional manner with the aid of DnaC helicase loader. ssDNA-loaded DnaB interacts with DnaG primase and DNA polymerase III holoenzyme, resulting in replisome (Arias-Palomo et al., 2019; Chodavarapu & Kaguni, 2016; Hayashi et al., 2020; O’Donnell et al., 2013; Sakiyama et al., 2017, 2018, 2022; Wegrzyn & Konieczny, 2023). Replisomes disassemble when the replication fork stalls in front of an obstacle, such as protein-DNA roadblock or single strand breaks. PriA recognizes the abandoned fork structure, which triggers reloading of DnaB with its partner proteins, such as PriB, PriC, DnaT, and DnaC (Heller & Marians, 2005a; Masai et al., 1994; Michel & Sandler, 2017; Windgassen et al., 2018). In addition, independently of PriA, PriC binds to the abandoned fork structure, triggering reloading of DnaB with the aid of DnaC, which is supported by Rep helicase depending on the fork structure (Heller & Marians, 2005b; Sandler et al., 1999). Also, homologous recombination mechanisms can rescue the abandoned replication fork during SOS responses (Asai & Kogomal, 1994; Kogoma, 1997; Né Dicte Michel et al., 2001). In contrast to these well characterized mechanisms at abandoned forks, little is known about what happens when replication initiation is impeded at oriC.

E. coli oriC comprises a Duplex-Unwinding Element (DUE) and a DnaA-Oligomerization Region (DOR) (Figure 1A) (Grimwade & Leonard, 2021; Kasho et al., 2023; Wegrzyn & Konieczny, 2023). The DUE contains 13-mer AT-rich sequence repeats known as L-, M-, and R-DUEs. The M- and R-DUEs are essential for stable DUE unwinding with the specific sequences (TT[A/G]T(T)) used for DnaA-ssDNA binding (Ozaki & Katayama, 2012; Sakiyama et al., 2017). The L-DUE containing the TTATT sequence promotes efficient DnaB loading by expanding the unwound oriC region (Sakiyama et al., 2022). The DOR is divided into three subregions: the Left-, Middle-, and Right-DORs, which contain asymmetric 9-mer DnaA binding sequences (DnaA box) with the consensus sequence TTA[T/A]NCACA (Figure 1A) (Noguchi et al., 2015; Ozaki & Katayama, 2012; Rozgaja et al., 2011; Sakiyama et al., 2017; Shimizu et al., 2016). The Left-DOR contains a cluster of unidirectionally aligned DnaA boxes including high-affinity DnaA box R1, and low-affinity boxes R5M, τ1−2, and I1-2, with an IHF-binding region between R1 and R5M boxes. The Right-DOR contains an oppositely oriented DnaA box cluster including high-affinity DnaA box R4 and low-affinity boxes I3 and C1-3. These DnaA box clusters form frameworks for the Left- and Right-DnaA subcomplexes, respectively. DnaA bound to the R2 box, which solely resides in the Middle-DOR, stabilizes these DnaA subcomplexes (Miller et al., 2009; Rozgaja et al., 2011; Shimizu et al., 2016). In addition, the AT-cluster (TATTAAAAAGAA) region, which connects to the L-DUE, stimulates DnaB loading in the absence of the Right-DnaA subcomplex (Sakiyama et al., 2022).

Figure 1

DnaA comprises four functional domains (Figure 1B). Domain I binds to multiple proteins such as DnaB helicases and the DnaA-assembly stimulator DiaA (Abe et al., 2007; Hayashi et al., 2020; Keyamura et al., 2007, 2009). Domain II is a flexible linker (Abe et al., 2007; Nozaki & Ogawa, 2008). Domain III contains AAA+ (ATPases Associated with various cellular Activities) motifs, which are involved in tight ATP/ADP binding, ATP hydrolysis, and domain III-III interactions (Duderstadt et al., 2011; Erzberger et al., 2006; Felczak & Kaguni, 2004; Kawakami et al., 2005; Ozaki et al., 2012). The arginine finger motif (Arg285) in domain III interacts with ATP bound to domain III of adjacent DnaA protomers, stimulating DnaA complex formation in a cooperative manner (Kawakami et al., 2005; Noguchi et al., 2015). In addition, H/B motifs (Val 211 and Arg245) in this domain bind to ssDNA in a sequence-specific manner (Ozaki et al., 2008; Sakiyama et al., 2017). Domain IV, which comprises the helix-turn-helix motif, recognizes the DnaA box (Fujikawa et al., 2003).

For replication initiation, ATP-DnaA molecules cooperatively oligomerize at oriC to form the Left- and Right-DnaA subcomplexes (Figure 1C). Formation of these complexes is stimulated by homotetrameric DiaA protein, which binds up to four DnaA molecules via domain I, including Phe46 (Keyamura et al., 2007, 2009). In concert with DNA superhelicity and thermal energy, the DUE undergoes initial unstable unwinding, which is stabilized through specific interaction between H/B motifs of domain III in the Left-DnaA subcomplex and the upper strand of M- and R-DUEs. This interaction is facilitated by sharp DNA bending by IHF, resulting in stable DUE unwinding (Figure 1C) (Ozaki et al., 2008; Sakiyama et al., 2017). When the single-stranded region is expanded to the L-DUE, the resultant single- stranded L-DUE binds the H/B motifs within the Right-DnaA subcomplex, maximizing the efficiency of the DnaB loading process (Sakiyama et al., 2022).

Subsequently, two DnaB helicases are loaded onto each ssDUE strand through interactions with DnaC and DnaA (Figure 1D). DnaC binding to DnaB changes the closed ring structure of the DnaB hexamer to an open spiral form to allow it to encircle ssDNA (Arias-Palomo et al., 2019; Nagata et al., 2020). Each Left- and Right-DnaA subcomplex binds a DnaB-DnaC complex via high-affinity interaction between DnaA Phe46 and DnaB Leu160 and guides the DnaB-DnaC complexes to unwound ssDUE strands via a low-affinity interaction between DnaA domain III His136 and DnaB (Figure 1B and 1D). Upon interaction with ssDNA, DnaC dissociates from DnaB, which enables migration of DnaB hexamer with helicase action. The pair of DnaB helicases loaded onto the ssDNA strands progress in opposite directions, forming replisomes with DnaG and Pol III holoenzyme (Chodavarapu & Kaguni, 2016; O’Donnell et al., 2013).

Replication restart pathways ensure fork progression of the entire chromosome under conditions that trigger replisome disassembly (Heller & Marians, 2005a; Lopper et al., 2007; Michel & Sandler, 2017). PriA helicase-dependent pathways predominantly facilitate DnaB reloading onto abandoned forks in vivo (Flores et al., 2002). PriA is a 3’-5’ DNA helicase and has specific affinity for forked DNA structures comprising one parental dsDNA and two newly synthesized sister dsDNA strands. This permits PriA recognition of the abandoned replication fork and the subsequent unwinding of the nascent lagging strand (Duckworth et al., 2023; Windgassen et al., 2018; Windgassen & Keck, 2016). The resultant unwound ssDNA associates with PriB and DnaT to promote reloading of DnaB helicase with the aid of DnaC (Duckworth et al., 2023; Heller & Marians, 2005a; Lopper et al., 2007). In the absence of PriB, PriC participates in a PriA-dependent pathway (Sandler et al., 1999). In addition, PriA is required for oriC/DnaA-independent chromosomal replication called stable DNA replication (SDR) (Masai et al., 1994), which is promoted by UV irradiation (inducible iSDR) and loss of rnhA or recG (constitutive cSDR) (Kogoma, 1997). In both mechanisms, replication is initiated at triple-stranded structures comprising a dsDNA-ssDNA hybrid (D-loop) or a dsDNA-ssRNA hybrid (R-loop) generated at specific chromosomal loci. These structures mimic the abandoned fork structure, thereby allowing helicase loading to occur similarly to that during PriA-dependent fork restart (Masai et al., 1994).

Independently of PriA, PriC can restart replication from abandoned forks through its interaction with the ssDNA region and SSB (Single-Stranded DNA Binding protein) (Figure 1E) (Heller & Marians, 2005a; Wessel et al., 2013, 2016). PriC consists of N-terminal and C-terminal domains, which are composed of α helices and are connected by a short linker. The PriC C-terminal domain remodels the SSB-ssDNA complex, which is a prerequisite for the recruitment of DnaB helicase (Heller & Marians, 2005a; Wessel et al., 2013, 2016). PriC also interacts with Rep helicase. When the length of the ssDNA gap within the abandoned fork is short, Rep helps expand the ssDNA gap, promoting PriC-mediated remodeling of the SSB-ssDNA complex (Heller & Marians, 2005a, 2005b; Nguyen et al., 2021). Notably, in addition to these functions at the abandoned forks, PriC is suggested to play a role in DnaA-oriC- dependent replication initiation under challenging conditions, as evidenced by the synthetic lethality between priC303::kan and a subgroup of temperature-sensitive dnaA mutant alleles (dnaA46 and dnaA508): i.e. priC303::kan mutants bearing dnaA46 or dnaA508, but not wild-type (WT) dnaA, can not grow even at 30°C (Hinds & Sandler, 2004). However, the mechanisms via which PriC rescues blocked replication initiation remain elusive.

In this study, we analyzed the PriC-dependent promotion of replication initiation from oriC in the context of various replication initiation stresses. We found that PriC stimulated replication initiation in dnaC2 cells, in which DnaB helicase loading at oriC is defective, indicating that it has a role in the rescue of blocked replication initiation. Moreover, PriC stimulated the growth of cells defective in DnaA-DnaB interactions in addition to the growth of cells with oriC sequence deletions that inhibit DnaB loading. Consistent with these results, in an in vitro reconstituted system, PriC stimulated DnaB loading when the DnaA-DnaB interaction was inhibited at oriC. Furthermore, we found that PriC did not stimulate initiation of cSDR, demonstrating that PriC functions specificity in the DnaA-oriC system. Taken together, we suggest that PriC rescues blocks in replication initiation by bypassing DnaB loading by oriC-DnaA complexes.

Results

PriC stimulates DNA replication initiation in dnaA46 and dnaC2 cells

The temperature-sensitive dnaA46 and dnaA508 mutants exhibit synthetic lethality with priC303::kan even at 30°C (Hinds & Sandler, 2004), suggesting that PriC can rescue blocks in DnaA- dependent replication initiation. However, the mechanism underlying PriC-promoted rescue of blocks in replication initiation remains a mystery. Based on previous reports demonstrating PriC binding to short oligo-ssDNA, SSB, and DnaB and its role in DnaB loading at stalled replication forks (Wessel et al., 2013, 2016), we considered the possibility that PriC permits bypass of DnaA-dependent stable DUE unwinding and/or DnaB loading at the oriC ssDUE. To genetically test these ideas, we analyzed the impact of PriC on the growth of a temperature-sensitive dnaC2 mutant, which is competent for ongoing replisome progression at 37–42°C, but defective in DnaB loading specifically at the oriC ssDUE (Withers & Bernander, 1998).

Colony formation of dnaA46 cells was moderately inhibited at 37℃ (Figure 2A). Consistent with a previous report, when crossed with the ΔpriC allele, colony formation of dnaA46 mutant cells but not WT dnaA cells was severely inhibited at 37℃ (Figure 2A). Unlike the dnaA46 priC303::kan mutant, dnaA46 ΔpriC cells formed colonies at 30°C, although colony growth was slightly retarded.

Figure 2.

This difference may stem from the different strain backgrounds. Notably, colony formation of dnaC2 cells at 35℃ was moderately inhibited, whereas it was severely inhibited in dnaC2 ΔpriC cells (Figure 2A). At 30°C, colony formation by dnaC2 cells and dnaC2 ΔpriC cells was similar. The observed requirement for PriC for colony formation by dnaC2 cells at 35℃ supports the idea that PriC assists in the DnaC-dependent DnaB loading step at oriC.

To investigate replication initiation from oriC, we used flow cytometry of synchronized cells bearing dnaA46 or dnaC2 with or without ΔpriC. In these experiments, cells were incubated at 42°C or 37°C for 80–90 min to synchronize the replication cycle at the pre-initiation step, after which incubation was continued at 30°C for 5 min to allow initiation and then 4 h in the presence of rifampicin and cephalexin to inhibit both replication initiation and cell division and allow replication run-out of the whole chromosome. The resultant number of chromosomes per cell, measured by flow cytometry, is known to correspond to the number of oriC copies per cell at the time of drug addition, an indicator of initiation activity.

A majority of the synchronized dnaA46 and dnaC2 cells had only a single oriC copy per cell after incubation at 42°C or 37°C regardless of the presence of ΔpriC (Figure 2B and 2C). In WT priC cells, after release of initiation at 30°C for 5 min, most cells had two oriC copies. However, in dnaA46 ΔpriC cells, only about half of the cells had two oriC copies after the release (Figure 2B), indicating moderate inhibition of initiation. In dnaC2 ΔpriC cells, the number of two-oriC cells was slightly lower than that of dnaC2 cells (Figure 2C). These observations are consistent with the colony formation data and with our hypothesis that PriC contributes to the rescue of blocked replication initiation by assisting DnaB loading onto the oriC ssDNA.

PriC rescues the cell growth of DnaA mutants defective in DnaB interactions

DnaA contains two regions for interactions with DnaB during DnaB loading at oriC ssDNA: one is a high-affinity region in domain I containing Phe46, and the other is a low-affinity region in domain III containing His136. To analyze the mechanism of PriC-promoted DnaB loading at oriC, we introduced a series of pING1 vector-based plasmids encoding WT DnaA (pKA234) or the DnaA F46A H136A double mutant (pFH) into dnaA46 cells with or without WT priC. Growth of dnaA46 cells containing pING1 was inhibited at 42℃, but the leaky expression of dnaA from pKA234 enabled the growth of these cells, as previously reported (Sakiyama et al., 2018) (Figure 3). Introduction of pFH moderately supported colony formation by dnaA46 cells at 42℃. By contrast, dnaA46 ΔpriC cells carrying pFH exhibited little growth at 42℃ (Figure 3). These results suggest that PriC is crucial for the growth of cells defective in specific DnaA-DnaB interactions, possibly because it can bypass the DnaA- DnaB interaction step required for DnaB loading onto oriC ssDNA.

Figure 3.

PriC rescues DNA replication inhibition by DiaA overexpression in vivo

DiaA, a homotetrameric protein, binds to multiple DnaA molecules and stimulates DnaA multimer assembly at oriC (Keyamura et al., 2007). The Phe46 residue in DnaA domain I is part of the DiaA-binding site and DnaB binding site. Oversupply of DiaA inhibits timely initiation of replication in cells growing in tryptone medium at 30°C, probably because of binding competition between DiaA and DnaB for DnaA domain I (Ishida et al., 2004; Keyamura et al., 2009). If PriC can bypass the DnaA-DnaB interaction requirement for DnaB loading onto oriC, PriC should suppress inhibition of initiation by DiaA oversupply. To test this possibility, we introduced pBR322 or its derivative encoding diaA (pNA135) into cells with or without WT priC (Figure 4). WT priC cells carrying pBR322 or pNA135 similarly formed colonies on LB agar plates containing ampicillin between 25°C and 42°C (Figure 4A and Figure 4-figure supplement 1). However, in ΔpriC cells carrying pNA135, but not pBR322, colony formation was moderately inhibited at 25°C (Figure 4A), supporting our conjecture that PriC can bypass the requirement for the DnaA-DnaB interaction.

Figure 4.

Figure 4-figure supplement 1.

Rep helicase stimulates PriC-mediated rescue of arrested replication forks containing nascent strands (Heller & Marians, 2005b, 2005a; Sandler, 2000). To test the contribution of Rep helicase to PriC-promoted rescue of blocked replication initiation, we introduced pBR322 or pNA135 into Δrep cells. Unlike ΔpriC cells, Δrep cells carrying pNA135 grew similarly to Δrep cells carrying pBR322 (Figure 4A), suggesting no involvement of Rep in the PriC-promoted rescue of blocked replication initiation. This is consistent with the features of unwound oriC, which has a fork structure similar to that of the replication fork but without nascent strands, and the specific role of Rep in rescuing abandoned replication forks with nascent strands in the PriC-dependent pathway. Rep is specifically required for the unwinding of the nascent leading strand of abandoned replication forks, which is needed to expand the ssDNA region required for entry of PriC (Heller & Marians, 2005b, 2005a).

We further analyzed replication initiation in ΔpriC cells carrying pNA135 at 30°C by flow cytometry (Figure 4B). Exponentially growing cells were further incubated for run-out replication after the addition of rifampicin and cefalexin. The volume (mass) of cells at the time of drug addition was also quantified. Under these conditions, WT priC cells bearing pBR322 mainly had eight or four oriC copies per cell with eight-oriC cells predominating (Figure 4B). The number of oriC copies per cell was similar to that in WT priC cells bearing pNA135, with only slight increase of asynchronous initiation, as shown by the slight increase in the number of cells with non-2ⁿ oriC copies, which is basically consistent with our previous data (Ishida et al., 2004). However, in ΔpriC cells bearing pNA135, severe inhibition of initiation occurred, as shown by the substantial decrease in the number of eight-oriC cells and the large increase of the number of cells with only one to three oriC copies (Figure 4B). This result supports the proposed function of PriC at oriC mentioned above. In addition, the observed asynchronous initiations in ΔpriC cells bearing pBR322 is worthy of attention. Indeed, the fraction of cells with non-2ⁿ oriC copies was higher among ΔpriC cells than among WT priC cells (Figure 4B), suggesting that even in normal cells, inhibition of replication initiation occurs at low frequency and requires PriC function to rescue replication initiation.

To test whether PriC rescues replication initiation inhibited by deletion of the diaA gene, we conducted similar flow cytometry analysis of ΔpriC, ΔdiaA, and ΔpriC ΔdiaA cells (Figure 4C). ΔdiaA cells showed severe asynchrony of replication initiation, as shown by the increase in the number of five- to-seven-oriC cells, which supports the stimulatory role of DiaA in DnaA-oriC complex formation (Figure 4C). However, ΔpriC ΔdiaA cells did not show more asynchronization of replication initiation than ΔdiaA cells, as shown by the similar asynchrony index (A.I.). These results suggest that PriC does not support or bypass DnaA-oriC complex formation.

PriC stimulates replication initiation in cells with oriC sequence truncations

Right DnaA-subcomplex plays a stimulatory role in DnaB helicase loading (Sakiyama et al., 2022). The loading of two DnaB hexamers at oriC ssDNA regions requires multiple steps that depend on the distinct functions of two DnaA subcomplexes (Figure 1): the Left-DnaA subcomplex stably unwinds M-R regions of the DUE and the Right-DnaA subcomplex expands the unwound region to the AT-L region, supporting efficient DnaB helicase loading (Sakiyama et al., 2017, 2022; Yoshida et al., 2023). The expanded ssDNA regions in this open oriC complex efficiently promote the loading of one DnaB hexamer onto the lower (A-rich) strand M-R region and a second DnaB hexamer onto the upper (T-rich) strand of the DUE (Fang et al., 1999; Sakiyama et al., 2022). In the absence of the Right-DnaA subcomplex, the AT region of the AT-L region assists in DnaB loading (Sakiyama et al., 2022; Stepankiw et al., 2009)

To determine whether PriC rescues defects in DnaB loading when the oriC sequence is altered to prevent Right-DnaA subcomplex formation, we first assessed the colony formation of cells lacking the Middle- and Right-DOR of oriC (Left-oriC) (Figure 5A). The Left-oriC mutant formed colonies at 37°C but showed severe growth defects at 25°C and 30°C (Figure 5B). At 37°C, thermal energy might be sufficient to expand DUE unwinding into the AT region and allow DnaB loading in the absence of the Right-DnaA subcomplex (Sakiyama et al., 2022), but not at 25°C or 30°C. By contrast, colony formation of the Left-oriC ΔpriC double mutant was markedly compromised at 37°C (Figure 5B) and the growth defects of the Left-oriC mutant 25°C and 30°C were aggravated. These findings support the idea that PriC functions to rescue defective processes in DnaB loading caused by deletions of the oriC sequence that prevent DnaA subcomplex formation.

Figure 5.

Next, similar experiments were performed using a mutant in which the R4 DnaA box within Right-DOR was substituted with a sequence (R4Tma) defective in binding of E. coli DnaA (Figure 5A and 5C) (Noguchi et al., 2015; Sakiyama et al., 2022). We reasoned that formation of the Right-DnaA subcomplex would be inhibited by the introduction of R4Tma and PriC would be required for robust initiation in this context. As in the Left-oriC strain, the R4Tma strain showed normal colony formation in the presence of WT priC. By contrast, in the R4Tma ΔpriC strain, colony formation was impaired moderately at 25°C and slightly at 30°C (Figure 5C), indicating that PriC can rescue the blocked replication initiation caused by the absence of the R4 box. The reduced colony formation of R4Tma ΔpriC double mutant cells was alleviated at higher temperatures, similar to the phenotype of the Left-oriC mutant.

Furthermore, we assessed replication initiation in R4Tma cells at 30°C using flow cytometry (Figure 5D). R4Tma cells grew slightly slower than WT oriC cells and showed clear inhibition of initiation; in contrast to the WT oriC strain, the R4Tma mutant cell population contained more four-oriC cells and fewer eight-oriC cells, and showed severe asynchronous initiation. Notably, these negative effects of the R4Tma mutation were amplified by deletion of priC, i.e., the number of one to three-oriC cells increased and the ori/mass ratio was further reduced.

We next analyzed the role of PriC in cells lacking the AT-L region of oriC (subATL oriC) (Figure 6A). L-DUE and the flanking AT region assist in DnaB helicase loading by stimulating oriC unwinding, which is essential in a strain defective in Right-DnaA subcomplex formation (Sakiyama et al., 2022). SubATL oriC cells and WT oriC cells formed colonies with similar efficiency on LB agar plates between 25°C and 37°C, regardless of the presence of ΔpriC (Figure 6B). However, flow cytometry analysis revealed that deletion of priC specifically inhibited initiation in subATL oriC cells (Figure 6C); the number of eight-oriC cells was lower while the number of four-to-seven oriC cells was higher in subATL oriC ΔpriC cells than in subATL oriC cells with WT priC. Taken together, these results are consistent with the idea that PriC restores initiation and DnaB loading at such truncated oriC sequences.

Figure 6.

PriC loads DnaB at oriC unwound by the DnaA complex in an in vitro reconstituted system

To analyze the mechanism of PriC rescue of blocked initiation at oriC, we assessed PriC activity in an in vitro reconstituted system for DnaB loading and DnaB-promoted DNA unwinding using the supercoiled circular form (form I) of the oriC plasmid pBSoriC and purified proteins, namely DnaA, N- terminally histidine-tagged DnaB (His-DnaB), DnaC, IHF, SSB, and gyrase. In this system, DnaA and IHF unwind the DUE and stimulate His-DnaB loading to the ssDUE region with the aid of DnaC. The loaded His-DnaB expands the ssDNA regions in concert with the ssDNA-binding activities of SSB. Concomitantly, the positive supercoiling generated through DNA unwinding by His-DnaB is resolved by DNA gyrase, resulting in the formation of plasmid DNA topoisomers (form I*). Form I* can be separated from form I by agarose gel electrophoresis (Sakiyama et al., 2018, 2022).

First, we assessed the effect of PriC on WT DnaA. WT DnaA stimulated form I* formation in the absence of PriC (Figure 7A, lanes 1 and 5) and increasing concentrations of PriC moderately inhibited WT DnaA-mediated form I* formation (Figure 7A, lanes 5 to 8), probably because of competition between DnaA and PriC for binding to DnaB. Next, we focused on the DnaA F46A H136A double mutant. The DnaA F46A and DnaA H136A mutant proteins alone are reported not to support form I* formation (Keyamura et al., 2009; Sakiyama et al., 2018). Consistently, the DnaA F46A H136A double mutant protein was inactive in form I* formation (Figure 7A, lane 9). However, when PriC was added to the assay, DnaA F46A H136A promoted form I* formation to a level comparable to that of WT DnaA in the presence of PriC (Figure 7A, lanes 5-12 and 7B), indicating that PriC rescues defective DnaB loading by DnaA F46A H136A-oriC complexes. These findings are consistent with the idea that PriC can bypass the strict reliance on DnaA-DnaB interactions for DnaB loading at oriC.

Figure 7.

Figure 7-figure supplement 1.

To corroborate this idea, we also determined whether PriC rescues blocked initiation resulting from unstable DUE unwinding. oriC complexes containing DnaA V211A cannot bind to the ssDUE and consequently fails in stable DUE unwinding (Ozaki et al., 2008). The DnaA V211A mutant was virtually inactive for form I * formation irrespective of the presence or absence of PriC (Figure 7A, lanes 13-16, and 7B), indicating that PriC cannot efficiently rescue a defect of DUE unwinding.

We further evaluated the DnaB loading activity of PriC in the presence of DiaA using a similar in vitro reconstituted system. Previously, DiaA was shown to stimulate oriC DUE unwinding, but to inhibit form I* formation, because DiaA-DnaA binding competitively inhibits DnaB-DnaA binding; DiaA and DnaB share the same binding site in DnaA domain I (Figures 7C lanes 1, 3, 5, and 7 and 7D) (Keyamura et al., 2009). PriC moderately stimulated form I* formation specifically in the presence of both DnaA and DiaA (Figures 7C lanes 7 and 8, and 7D). These results further support the idea that PriC can bypass the specific requirement for the DnaA-DnaB interaction in DnaB loading at oriC. DiaA bound to oriC-DnaA complexes might be a physical obstacle reducing efficiency of DnaB loading by PriC.

In addition, we analyzed the DUE unwinding activities of DnaA mutants using the in vitro reconstituted system. As previously reported, DnaA V211A is largely inactive in DUE unwinding, but DnaA F46A H136A unwinds the DUE at a level comparable to that of WT DnaA (Figure 7-figure supplement 2A and 2B). PriC did not stimulate DUE unwinding irrespective of the presence of WT DnaA (Figure 7-figure supplement 2C and 2D). These results demonstrate the functional specificity of PriC in rescuing failed DnaB loading at oriC ssDNA and are consistent with the PriC mechanism mentioned above.

Role for PriC in cSDR

Finally, to analyze the role of PriC in other types of replication initiation, we examine whether PriC contributes to the initiation of cSDR, which does not require dnaA or oriC and is activated in cells lacking rnhA or recG. Although the PriA-PriB-DnaT primosome complex has previously been reported to be involved in the initiation of cSDR (Heller & Marians, 2006), it is unclear whether PriC has any role in this type of initiation.

Unlike dnaA46 cells, dnaA46 rnhA::cat double mutant cells formed colonies even at 40°C (Figure 8A), indicating that these cells were engaged in cSDR, as previously reported (Hinds & Sandler, 2004). The formation of colonies by ΔpriC dnaA46 rnhA::cat triple mutant cells was severely inhibited at 40°C (Figure 8A), suggesting that PriC contributes to the growth of dnaA46 rnhA::cat mutant cells.

Figure 8.

To determine whether PriC contributes to the initiation of dnaA46-oriC replication or cSDR, we calculated the relative ratios of oriC (84.6 min) and ter (32.4 min) copy numbers to the refence locus yapB (50.5 min) copy number using real-time quantitative PCR (qPCR). To calculate the copy number of the chromosomal ter region, the preferential initiation site of cSDR (Maduike et al., 2014), dnaA46 rnhA::cat cells growing exponentially at 30°C were shifted to 40°C for 90 min and the copy number before and after incubation at 40°C was calculated. Consistent with the colony formation data (Figure 8A), the ter copy number ratio of the dnaA46 rnhA::cat double mutant increased after 40°C incubation, whereas the copy number ratio of the dnaA46 mutant expressing WT rnhA remained the same. The ter copy number ratio of the dnaA46 rnhA::cat ΔpriC triple mutant strain was also increased, indicating that cSDR could occur even in the absence of PriC.

To monitor initiation of DnaA-oriC replication, we determined the oriC copy number ratio (Figure 8B). At the permissive temperature, the oriC copy number ratio of the dnaA46 ΔpriC double mutant strain was lower than that of the dnaA46 single mutant, confirming the importance of PriC for replication initiation in dnaA46 cells, as shown in Figure 2. In the dnaA46 rnhA::cat double mutant expressing WT priC, the oriC copy number ratio was reduced, which is explained by activation of cSDR and increase of the relative copy number of the reference yapB locus (50.5 min chromosomal map position). Together, these results suggest the idea that PriC stimulates replication fork progression during cSDR.

Discussion

In growing cells, replication initiation can be blocked under challenging conditions. However, the mechanisms that rescue blocked initiations are largely unknown. To elucidate such mechanisms, we investigated PriC-promoted replication initiation under various challenging conditions. We showed that PriC was necessary for the optimal growth of dnaC2 cells at the semi-restrictive temperature and for replication initiation even at the permissive temperature (Figure 2), suggesting that PriC contributes to replication initiation rescue in dnaC2 cells by facilitating DnaB helicase loading at oriC. Furthermore, we observed that PriC was necessary for the optimal growth of cells in which the DnaA-DnaB interaction was inhibited by DiaA overexpression or DnaA F46A H136A double mutations (Figures 3 and 4). In addition, PriC was stimulatory for the colony formation and replication initiation bearing oriC mutations that impair efficient DnaB loading (Figures 5 and 6). These results, together with results of form I* formation assays (Figures 7), suggest that PriC bypasses the need for the DnaA-DnaB interaction by binding to unwound oriC DNA and recruiting DnaB to facilitate its loading onto ssDNA (Figure 9). This mechanism is similar to that proposed for the rescue of abandoned replication forks by PriC (Figure 1E).

Figure 9.

The finding that PriC failed to stimulate form I* formation in the presence of DnaA V211A, which lacks DUE unwinding activity (Figures 7), provides mechanistic insight into PriC-dependent replication initiation at oriC. We propose that PriC-dependent replication initiation possibly begins with the recognition of a stably unwound ssDNA region by PriC (Figure 9). Consistent with this perspective, PriC ssDNA-binding activity and ssDNA regions have been reported to be important for PriC-promoted replication fork rescue (Heller & Marians, 2006; McMillan & Keck, 2024). The failure of PriC to stimulate replication initiation in ΔdiaA cells (Figure 4C) is also potentially explained by the reduced DUE unwinding caused by destabilization of the DnaA-oriC complex in ΔdiaA cells. Based on this, we suggest that the stably unwound ssDNA region in oriC serves as a distinctive feature for PriC- promoting DnaB loading at oriC for replication initiation rescue, and distinguishes it from PriA- dependent replication restart, which requires the structure of the replication fork DNA consisting of sister dsDNA strands (Heller & Marians, 2005a; Windgassen et al., 2018).

In this study, we found that cells with DnaA F46A H136A double mutations grew well, but only when the cells expressed WT PriC, whereas the growth of cells expressing the DnaA H136A single mutant was severely inhibited even when the cells expressed WT PriC (Sakiyama et al., 2018). The His136 residue is located within the weak, secondary DnaB interaction region in DnaA, and is crucial for DnaB loading onto oriC ssDNA. Although domain I in DnaA H136A can stably tether DnaB-DnaC complexes to DnaA complexes on oriC (Sakiyama et al., 2018), the complexes fail to load DnaB onto oriC ssDNA even in the presence of PriC. It is possible that the interaction between PriC and DnaB is inhibited by DnaB binding to DnaA domain I. Conversely, the PriC-DnaB interaction may inhibit the interaction between DnaA domain I and DnaB, as suggested by the inhibitory effect of PriC on DnaB loading in vitro in the presence of WT DnaA (Figure 7). However, this inhibitory effect of PriC was not observed in vivo, suggesting that PriC-DnaB binding could involve still unknown factors and regulatory mechanisms.

PriC supported cSDR-dependent growth in dnaA46 rnhA::cat double mutant cells without stimulating the initiation of cSDR (Figure 8). In this mutant, cSDR predominantly initiates from the ter region, resulting in head-on collisions between the replisome and transcription complexes in rrn operons (Maduike et al., 2014). This suggests that in cSDR cells PriC specifically rescues abandoned forks including those generated during head-on conflicts between the replisome and transcription complexes.

Bacillus subtilis (B. subtilis), which expresses a PriA homolog but not a PriC homolog, employs a PriA-independent fork rescue pathway (Bruand et al., 2001). In this pathway, B. subtilis DnaC (BsuDnaC) helicase is reloaded using the helicase loader DnaI together with B. subtilis DnaB (BsuDnaB) and DnaD coloaders. The ssDNA-binding activity of BsuDnaB has a crucial role in this PriA-independent pathway (Bruand et al., 2005). SsDNA-bound BsuDnaB remodels the SSB-ssDNA complex with DnaD and recruits BsuDnaC. These functions of BsuDnaB are similar to the proposed functions of PriC in replication fork rescue in E. coli. In addition, BsuDnaC helicase is recruited to oriC by concerted actions of B. subtilis DnaA and the DnaD-BsuDnaB complex, which has an essential role in chromosome replication initiation (Jameson & Wilkinson, 2017), suggesting a similar role of BsuDnaB to PriC in E. coli cells expressing DnaA F46A H136A. These similarities between PriC and BsuDnaB in these two evolutionarily distant bacterial species suggest that the mechanism of PriC- promoted helicase loading is conserved among bacterial species despite the absence of sequence conservation of PriC and BsuDnaB homologs.

Based on our results, we propose the abnormal competition between DnaB and DiaA for DnaA domain I could represent a form of intrinsic replication initiation stress in bacteria with conserved DiaA homologs. This type of stress could also occur in ε-proteobacterial species, such as Helicobacter pylori, because they express HobA, a DiaA-functional homolog (Natrajan et al., 2007; Zawilak-Pawlik et al., 2011). Similarly, YfdR, encoded by E. coli prophages, inhibits replication initiation by competing with DnaB for binding to DnaA domain I (Noguchi & Katayama, 2016), suggesting that such extraneously introduced inhibitors could trigger replication initiation stress.

Even in human, ORC1, ORC2, and ORC5, the essential components of the eukaryotic replication initiation complex, are not essential in some cancer cell-lines (Shibata & Dutta, 2020; Struhl et al., 2016), suggesting that mechanisms of replication initiation rescue may also operate beyond the bacterial kingdom. Therefore, further investigation of the initiation rescue processes and the factors involved in diverse organisms from bacteria to human will be important for a full understanding of the common principles and diverse mechanisms that ensure robust initiation of chromosomal DNA replication.

Materials and methods

Plasmids, proteins, and strains

The plasmids used in this study are listed in Key Resource Table. pKA234, pKW44-1, pNA135, pBSoriC, pET22b(+)-priC, and pTKM601 were described previously (Aramaki et al., 2015; Ishida et al., 2004; Kawakami et al., 2005; Keyamura et al., 2009; Kubota et al., 1997; Ozaki et al., 2008). For the construction of pFH, an alanine substitution was introduced into pH136A with specific mutagenic primer sets using the QuikChange site-directed mutagenesis protocol (Stratagene [Agilent], Agilent, La Jolla, CA, United States) as described previously (Keyamura et al., 2007; Sakiyama et al., 2018). WT DnaA and its derivative proteins were overproduced in E. coli strain KA450 from pKA234, pKW44-1, or pFH and purified as described previously (Noguchi et al., 2015; Ozaki et al., 2008; Sakiyama et al., 2017).

PriC protein was prepared as previously reported with minor modifications (Aramaki et al., 2013). Briefly, PriC protein was overproduced in E. coli strain BL21-codonPlus-RIL by inducing its expression from pET22b(+)-priC using 0.2 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG). The resulting cells were suspended in chilled buffer A (50 mM HEPES-NaOH [pH7.0], 10% sucrose, 1 mM EDTA, 2 mM dithiothreitol (DTT), and 1 mM PMSF) and disrupted by sonication. Proteins in the soluble fraction were precipitated with 0.24 g/mL ammonium sulfate, resuspended in a separation buffer (50 mM imidazole [pH 7.0], 20% glycerol, 2 mM DTT, 1 mM EDTA, and 40 mM ammonium sulfate) and loaded onto a 1 mL HiTrap® SP HP column (Cytiva, Uppsala, Sweden). Bound proteins were eluted with a linear gradient from 0 to 1 M sodium chloride.

All E. coli strains used in this study are listed in Key Resource Table. KYA018, MIT125, MIT162, SA103, NY20 and NY21 were described previously (Kasho & Katayama, 2013; Noguchi et al., 2015; Noguchi & Katayama, 2016). Strains bearing mutant oriC (NY20L and NY20ATL) were constructed using the λRed site-directed recombination system as previously described (Noguchi et al., 2015; Sakiyama et al., 2017, 2022). NY20-frt strain and oriC mutant strains (NY21-frt, NY20L-frt, and NY20ATL-frt) were constructed by eliminating kan from the NY20 strain and strains NY21, NY20L, and NY20ATL. To remove the kanamycin-resistant cassette (kan) from oriC, FLP recombinase encoded on pCP20 was used. Elimination of kan was verified by checking sensitivity to 50 μg/ml kanamycin in LB agar plates. ΔpriC::frt-kan from JW0456-KC and Δrep::frt-kan from JW5604-KC were introduced into MG1655 by P1 transduction, resulting in the construction of KRC002 and FF001, respectively. For construction of other ΔpriC::frt-kan strains (KRC004, KRC005, NY20-priC, NY21-priC, NY20L-priC, NY20ATL-priC), P1 phage lysates prepared from KRC002 were used for transduction of strains MIT125, KYA018, NY20-frt, SYM21-frt, NY20L-frt, and NY20ATL-frt. Transductants were screened on LB agar plates containing 50 µg/mL kanamycin. ΔdiaA::frt-kan or rnhA::cat derivatives (KRC006, MIT125c, and KRC004c) were constructed using a similar protocol except that P1 phage lysates from SA103 (ΔdiaA::frt-kan) or MIT162 (rnhA::cat) were used. KRC003 was generated by removal of kan from KRC002.

Buffers

Buffer P contained 60 mM HEPES–KOH (pH 7.6), 0.1 mM zinc acetate, 8 mM magnesium acetate, 30% [v/v] glycerol, and 0.32 mg/mL bovine serum albumin (BSA). Buffer N contained 50 mM HEPES–KOH (pH 7.6), 2.5 mM magnesium acetate, 0.3 mM EDTA, 7 mM DTT, 0.007% [v/v] Triton X-100, and 20% [v/v] glycerol. Form I* buffer contained 20 mM Tris-HCl (pH 7.5), 125 mM potassium glutamate, 10 mM magnesium acetate, 8 mM DTT, and 0.5 mg/mL BSA.

DUE unwinding assay

DUE unwinding assays were performed essentially as described with minor modifications (Sakiyama et al., 2022; Yoshida et al., 2023). Briefly, DnaA was incubated with ATP in buffer N on ice for 3 min to generate ATP-DnaA. The indicated amount of PriC and ATP-DnaA or its mutant derivatives was incubated for 9 min at 30℃ in 10 μL buffer P containing 5 mM ATP, 125 mM potassium glutamate, 1.6 nM pBSoriC, and 42 nM IHF, followed by further incubation with 1.5 units of P1 nuclease (Wako) for 5 min. The reaction was stopped by the addition of 1% SDS and 25 mM EDTA, and DNA was purified by phenol-chloroform extraction and ethanol precipitation. One-third of each purified DNA was digested with AlwNI (NEB), which yielded 2.6 kb and 1.0 kb fragments after DUE unwinding of pBSoriC. The resultant DNA fragments were analyzed by 1% agarose gel electrophoresis with 1xTris- acetate-EDTA buffer for 30 min at 100 V, followed by ethidium bromide staining. Gel images were taken using the GelDoc GO imaging system (Bio-Rad Laboratories, Hercules, CA), and products derived from unwound plasmids were quantified using ImageJ software.

Form I* assay

This assay was performed as previously described with minor modifications (Noguchi et al., 2015; Sakiyama et al., 2022). The indicated amount of ATP-DnaA was incubated for 15 min at 30℃ in 12.5 μL of Form I* buffer containing 3 mM ATP, 1.6 nM pBSoriC, 42 nM IHF, 400 nM His-DnaB, 400 nM DnaC, 76 nM GyrA, 100 nM His-GyrB, and 760 nM SSB. The reaction was stopped by the addition of 0.5% SDS, and DNA was purified by phenol-chloroform extraction. Samples were analyzed by 0.65% agarose gel electrophoresis with 0.5xTris-borate-EDTA buffer for 15 h at 23 V, followed by ethidium bromide staining.

Flow cytometry analysis

Flow cytometry analysis was performed essentially as described (Kasho et al., 2014). Briefly, cells were grown at 30℃ in LB medium until the absorbance of the culture (A₆₀₀) reached 0.1. Portions of the cultures were diluted a thousand-fold into 5 ml LB medium and incubated at 30℃ until the absorbance of the culture (A600) reached 0.1. The remaining portions were further incubated to determine the doubling time (Td) by measuring A₆₀₀ every 20 min. At A₆₀₀ 0.1, aliquots of the cultures were fixed in 70% ethanol to analyze cell mass using the Multisizer 3 Coulter counter (Beckman Coulter, Brea, CA). The remaining cultures were further incubated for 4 h with 0.3 mg/mL rifampicin and 0.01 mg/mL cephalexin to allow run-out replication of chromosomal DNA. The resultant cells were fixed in 70% ethanol. After DNA staining with SYTOX Green (Thermo Fisher Scientific, Waltham, MA), cellular DNA contents were analyzed on a FACS Calibur flow cytometer (BD Bioscience, Franklin Lakes, NJ). The number of the origins/cell (ori/cell) was determined from the histograms of flow cytometry analysis. The number of ori/cell was divided by the mean cell mass determined by cell mass analysis, resulting in ori/mass ratios. The numbers of cells containing non-2ⁿ copy number of oriC were divided by the number of cells containing 2ⁿ copy number of oriC to calculate the asynchrony index (A.I.). For the analysis of cells bearing plasmids, LB medium was supplemented with 100 μg/mL ampicillin.

Replication initiation frequency test using synchronized cultures

MIT125 (dnaA46, tnaA::Tn10) cells or its ΔpriC derivative KRC004 were grown at 30℃, a permissive temperature, in LB medium until the absorbance of the culture (A₆₀₀) reached 0.04, followed by further incubation for 90 min at 42℃, a restrictive temperature. Aliquots of the cultures were fixed in 70% ethanol as “Synchronization” samples. The remaining cultures were further incubated for 5 min to initiate DNA replication, followed by further incubation for 4 h at 42℃ with 0.3 mg/mL rifampicin and 0.01 mg/mL cephalexin to allow run-out replication of chromosomal DNA. The resultant cells were fixed in 70% ethanol as “5 min release” samples. After DNA staining with SYTOX Green (Life Technologies), cellular DNA contents were analyzed on a FACS Calibur flow cytometer (BD Bioscience). Ratios of the origins/cell (ori/cell) were determined from the histograms of flow cytometry analysis.

For synchronization of dnaC2 mutant cells (KYA018 and KRC005), cells were grown in LB medium at 30℃ until the A₆₀₀ reached 0.04, followed by further incubation for 80 min at 37℃, at the restrictive temperature. The resulting synchronized cultures were released for 5 min at 30℃ and incubated in the presence of 0.3 mg/mL rifampicin and 0.01 mg/mL cephalexin at 30℃ to allow run-out of chromosomal DNA replication.

qPCR for analysis of cSDR

Cells of MIT125 (dnaA46, tnaA::Tn10), MIT125c (dnaA46, tnaA::Tn10, rnhA::cat) and its ΔpriC derivatives KRC004 (dnaA46, tnaA::Tn10, ΔpriC::frt-kan) and KRC004c (dnaA46, tnaA::Tn10, rnhA::cat, ΔpriC::frt-kan) were grown at 30℃ in LB medium until the absorbance of the culture (A₆₀₀) reached 0.04 and aliquots of the cultures were withdrawn. The remaining samples were further incubated for 90 min at 40℃ and aliquots of the cultures were withdrawn. These samples were boiled for 5 min at 95℃ and the genome DNA was extracted. The levels of oriC (84.6 min), ter (32.4 min), and yapB (50.5 min) were quantified by real-time qPCR using TB Green Premix Ex TaqII (Tli RNaseH Plus) (TaKaRa, Shiga, Japan) and the following primers: ORI_1 and KWoriCRev for oriC, qoriK fw and qoriK rev for ter, STM419 and STM420 for yapB.

Acknowledgements

We are grateful to Dr. Yoshio Abe for discussion and priC-overproducing plasmid, to Drs. Yusuke Akama and Yukari Sakiyama for initial exploratory study related to this work, and to NRBP-E. coli at NIG for bacterial strains.

Funding

Japan Society for the Promotion of Science (JSPS KAKENHI) [JP17H03656, JP20H03212 and JP23K27131].

Key Resources Table