Replication of Vibrio cholerae chromosome I in Escherichia coli: dependence on dam methylation

Abstract

We successfully substituted Escherichia coli's origin of replication oriC with the origin region of Vibrio cholerae chromosome I (oriCI(Vc)). Replication from oriCI(Vc) initiated at a similar or slightly reduced cell mass compared to that of normal E. coli oriC. With respect to sequestration-dependent synchrony of initiation and stimulation of initiation by the loss of Hda activity, replication initiation from oriC and oriCI(Vc) were similar. Since Hda is involved in the conversion of DnaA(ATP) (DnaA bound to ATP) to DnaA(ADP) (DnaA bound to ADP), this indicates that DnaA associated with ATP is limiting for V. cholerae chromosome I replication, which similar to what is observed for E. coli. No hda homologue has been identified in V. cholerae yet. In V. cholerae, dam is essential for viability, whereas in E. coli, dam mutants are viable. Replacement of E. coli oriC with oriCI(Vc) allowed us to specifically address the role of the Dam methyltransferase and SeqA in replication initiation from oriCI(Vc). We show that when E. coli's origin of replication is substituted by oriCI(Vc), dam, but not seqA, becomes important for growth, arguing that Dam methylation exerts a critical function at the origin of replication itself. We propose that Dam methylation promotes DnaA-assisted successful duplex opening and replisome assembly at oriCI(Vc) in E. coli. In this model, methylation at oriCI(Vc) would ease DNA melting. This is supported by the fact that the requirement for dam can be alleviated by increasing negative supercoiling of the chromosome through oversupply of the DNA gyrase or loss of SeqA activity.

FIG. 1.

Alignment of the E. coli minimal oriC with the corresponding region from V. cholerae chromosome I. The AT-rich sequence and the three 13-mer repeats L, M, and R found in E. coli (5) are indicated above the alignment. The 6-mer (A/T)GATCT boxes (80) are underlined. Other DnaA binding sites, i.e., R-boxes (53), I-boxes (52), and τ-boxes (61), are shown as boxed regions. Dam methylation sites (GATC) are shaded gray. The experimentally defined binding sites for integration host factor (IHF) (22) and factor for inversion stimulation (FIS) (65) in E. coli are indicated, and bases that match the consensus sequence are in boldface type. The single base difference between oriCI_Vc and oriCI_Vc* (see Materials and Methods) in the minimal origin region is shown below the two sequences. A gap introduced to maximize alignment of the two sequences is indicated by a dash in the sequence. Nucleotides that are identical in the two sequences are indicated by an asterisk below the two sequences.

FIG. 2.

Copy number of oriCI-based minichromosomes. Total DNA was isolated from E. coli strain XL1-Blue containing the indicated minichromosomes. Cells were grown exponentially at 37°C in AB minimal medium supplemented with glucose and Casamino Acids and containing 50 μg/ml of kanamycin. Individual DNA samples were digested with BamHI plus XhoI before Southern blot hybridization was performed. Two probes were used simultaneously; the two probes were homologous to sequences specific for terC on the chromosome and the kanamycin resistance gene carried by the minichromosomes.

FIG. 3.

Structure of genes in the origin regions of the strains used in this study. (A) E. coli wild type (Wt). (B) The aadA gene (GenBank accession no. AAC33912) encoding streptomycin adenylyltransferase flanked by transcription and translation termination signals (the ΩSm cassette [66]) is inserted upstream of the E. coli gidA promoter as defined in reference . (C) The E. coli oriC was replaced by the oriCI_Vc region comprising the intergenic region between the V. cholerae mioC and gidA gene homologues and the sequence encoding the first 20 amino acids of the V. cholerae gidA gene homologue in combination with the ΩSm cassette. (D) Same as panel C but with two point mutations. The precise locations of the point mutations are described in the text.

FIG. 4.

Cell cycle parameters for wild-type (Wt) (E. coli MG1655), ΩSm, oriCI_Vc, and oriCI_Vc* strains. Wt, ΩSm, oriCI_Vc, and oriCI_Vc* cells were grown at 37°C in minimal medium supplemented with glycerol, glucose, or glucose plus Casamino Acids (CAA). Cells were treated with rifampin and cephalexin (Materials and Methods) prior to flow cytometric analysis.

FIG. 5.

Loss of Hda activity stimulates replication initiation from oriCI_Vc. ΩSm cells and hda::cat derivatives of ΩSm, oriCI_Vc, and oriCI_Vc* cells were grown at 37°C in minimal medium supplemented with glucose (Glu) plus Casamino Acids (CAA). The cells were treated with rifampin and cephalexin (Materials and Methods) prior to flow cytometric analysis.

FIG. 6.

Synchronous replication initiation from oriCI_Vc is dependent on seqA and dam gene products. Cells were grown at 37°C in minimal medium supplemented with Glu plus CAA. Panels A, C, E, G, H, I, and J show cells treated with rifampin and cephalexin prior to flow cytometric analysis, whereas the cells in panels B, D, and F were in the exponential growth phase. seqA (A and B), ΩSm seqA (C and D), oriCI_Vc seqA (E and F), damX::mini-Tn10 (G), ΩSm damX::mini-Tn10 (H), oriCI_VcdamX::mini-Tn10 (I), and oriCI_Vc* damX::mini-Tn10 (J) strains were studied. Insertion of mini-Tn10 in damX reduces the transcription of the dam gene to approximately 10% of the level in wild-type cells (42).

FIG. 7.

Dam-deficient oriCI_Vc* cells contain suppressor mutations. The dam16::Km^r allele was P1 transduced into oriCI_Vc* (A) and wild-type (wt) (B) cells. Four independent oriCI_Vc* dam cells (from panel A) were transduced back to Dam⁺ by a two-step procedure (details in Materials and Methods) (46). By this procedure, putative dam suppressor mutations (dsm mutations) will be present in an otherwise Dam⁺ background. The dam16::Km^r allele was subsequently transduced back into the oriCI_Vc* (hsm?) cells. This transduction resulted in a high number of transductants, indicating that all four oriCI_Vc* dam clones tested contained secondary mutations to compensate for loss of Dam activity. The same dam16::Km^r P1 lysate was used for all strains, and the same number of cells was plated on selective media and incubated for 24 h at 37°C prior to inspection.

FIG. 8.

Depletion of Dam methylase leads to oriCI_Vc initiation arrest. Dam methylase was depleted from wt, ΩSm, oriCI_Vc, and oriCI_Vc* cells as described in Materials and Methods. Samples were incubated with rifampin and cephalexin prior to flow cytometric analysis. (A) DNA histograms of oriC and oriCI_Vc after incubation in the presence of IPTG for the indicated time period. Cells were treated with rifampin and cephalexin prior to flow cytometric analysis. (B) Two-parameter histograms showing DNA content versus cell size after incubation with IPTG for 0 and 7 h.

FIG. 9.

Dam methylation is dispensable for oriCI_Vc function in cells with increased negative supercoiling. (A) The dam16::Km^r allele was transduced into ΩSm seqA and oriCI_Vc seqA cells derived from E. coli MG1655. (B) The dam16::Km^r allele was transduced into ΩSm and oriCI_Vc cells carrying the gyrAB genes under the control of the IPTG-inducible P_A1lacO1 promoter. (C) The dam16::Km^r allele was transduced into ΩSm seqA and oriCI_Vc seqA cells carrying the gyrAB genes under the control of the IPTG-inducible P_A1lacO1 promoter. Transductants were plated on LB agar plates containing kanamycin (A) or kanamycin, tetracycline, and the indicated IPTG concentration (B and C) and were inspected after 22 h of incubation at 42°C. The same dam16::Km^r P1 lysate was used for all strains, and the same number of cells was plated.