Multipartite regulation of rctB, the replication initiator gene of Vibrio cholerae chromosome II

Abstract

Replication initiator proteins in bacteria not only allow DNA replication but also often regulate the rate of replication initiation as well. The regulation is mediated by limiting the synthesis or availability of initiator proteins. The applicability of this principle is demonstrated here for RctB, the replication initiator for the smaller of the two chromosomes of Vibrio cholerae. A strong promoter for the rctB gene named rctBp was identified and found to be autoregulated in Escherichia coli. Promoter activity was lower in V. cholerae than in E. coli, and a part of this reduction is likely to be due to autorepression. Sequences upstream of rctBp, implicated earlier in replication control, enhanced the repression. The action of the upstream sequences required that they be present in cis, implying long-range interactions in the control of the promoter activity. A second gene specific for chromosome II replication, rctA, reduced rctB translation, most likely by antisense RNA control. Finally, optimal rctBp activity was found to be dependent on Dam. Increasing RctB in trans increased the copy number of a miniplasmid carrying oriCII(VC), implying that RctB can be rate limiting for chromosome II replication. The multiple modes of control on RctB are expected to reduce fluctuations in the initiator concentration and thereby help maintain chromosome copy number homeostasis.

FIG. 1.

Identification by deletion mapping of a region containing a promoter for the initiator gene rctB. The top line shows the region involved in chromosome II replication. It consists of rctB, the origin (oriCII_VC), and elements of negative control of replication. The region has two kinds of repeats, 11-mers (open triangles) and 12-mers (filled triangles), and these were used as guides for choosing deletion end points (next five lines). The coordinates are as published previously (17). All fragments ended at coordinate 1157 and included the first eight codons of the rctB ORF. They were cloned in front of the lacZ reporter in a promoter-cloning vector (pMLB1109), and their promoter activities were measured after the resultant plasmids were transferred to E. coli (DH5Δlac) and V. cholerae (CVC250). The arrows indicate the orientation in front of the lacZ gene. The reporter (β-galactosidase) activities were measured and presented in Miller units as described previously (20). Values presented are averages from at least two independent transformants, each assayed in duplicate.

FIG. 2.

Precise location of rctBp. (A) Primer extension results showing the sequence around the start site of the rctB message. The starting base is shown in bold. (B) The predicted −10 and −35 boxes, assuming that +1 is at coordinate 1092, are bracketed. Also shown are the predicted start codon (in bold at 1134) and a putative SD sequence (underlined) of the rctB ORF. The promoter region has four heptameric sequences (boxes) containing GATC that are partially homologous to 11- and 12-mer repeats. Their sequence relationships are shown in the box at the bottom. Also shown are two potential ToxT binding sites (gray arrows) overlapping the −35 and −10 boxes. (C) Promoter activities when the putative −10 and −35 boxes of rctBp were mutated. The mutated sequences are shown above the boxes in panel B.

FIG. 3.

Transcriptional autorepression of rctBp. (A) The promoter activities from two different fragments were determined: when they were present in transcriptional fusion vector pMLB1109, the resultant plasmids were called pDP309 and pDP310 (Fig. 1), and when in translational fusion vector pMLB1034, the resultant plasmids were called pDP311 and pDP312. To study autorepression, various levels of RctB protein were supplied from pTVC11 (pTVC12+rctB) where the protein was under arabinose-inducible promoter P_BAD present in vector pTVC12. The host was BR8706 in these experiments, with a constitutive source of arabinose transporter. (B) To show specificity of promoter repression by RctB, activity of another fusion, repAp-lacZ (present in pALA326), was tested. RepA was supplied from pALA169, where repA is constitutively made. pST52 is the vector used to clone repA that generated pALA169. (C) The arabinose induction of P_BAD was confirmed by measuring the activity of the luciferase gene present downstream of rctB in pTVC11. pTVC12 is identical to pRFG110 except that rctB and the downstream luciferase gene replaced the lacI-yfp fusion gene. (D) RctB production after P_BAD induction was measured by immunoblotting with antibody against RctB. Extracts from equal ODs of cells were added to all lanes, except that the ODs of V. cholerae cells were two times higher (lanes 14 and 15). Purified RctB (5, 10, and 20 ng) (arrow) were added to extracts from vector (pTVC12)-carrying cells for reference purposes (lanes 1 to 3). The band inferred as RctB (arrow, lanes 14 and 15) comigrated with purified RctB when it was mixed with the V. cholerae extract (data not shown), confirming the identity of the protein. Samples were loaded in duplicates (lanes 4 to 15).

FIG. 4.

Search for additional promoters of rctB. The fragments shown were fused to lacZ of the promoter-cloning vector pMLB1109, as in Fig. 1. The region containing rctBp is deleted in all but pDP309. In the +RctB column, the protein was supplied from pTVC11 after adding 0.2% arabinose to the culture; otherwise, the experiments were done as described in the legend to Fig. 3. In the −RctB column, the vector pTVC12 replaced pTVC11.

FIG. 5.

Effect of rctA on rctB expression. The rctA effect was determined either alone or in conjunction with RctB. (A) A transcriptional rctBp-lacZ fusion as in pDP336 (Fig. 1) was used, but the fusion was present in an integrated λ prophage in CVC444 (Table 1). RctB was supplied either from an inducible source, pTVC11, or constitutively from pTVC13 (a pET28a derivative, but the cells did not have a source of T7 polymerase) when rctA was also present in the same cell. When present, rctA was under P_BAD control as in pDP354. (B) Same as panel A except the fusion was translational as in pDP351. (C) Same as panel B except rctBp was replaced with repAp and the resultant strain called BR3239. pALA162 was used to supply RepA to confirm the repressibility of repAp. (D) The extent of complementarity between the rctA message and the leader of rctB message is shown by vertical lines. The first 48 nucleotides of the leader of the rctB mRNA are shown in the top line. The predicted sequence of the 5′ end of rctA message as made from the P_BAD promoter present in pDP354 is shown below. The predicted start codon of rctA ORF is shown in gray (coordinate 246), and the vector-derived 5′ end sequences are in italics. The chromosomal sequences covering rctA present in pDP354 span the coordinates 845 to 192.

FIG. 6.

Relationship of RctB concentration and the copy number of a minichromosome carrying oriCII_VC. The origin region was present in pTVC31, and various levels of RctB were supplied from pTVC11 by changing the arabinose concentration. A constant volume of a culture of pNEB193 was added to each sample before the start of plasmid DNA isolation to normalize for any loss of DNA during the isolation procedure. The DNAs were linearized using SalI before running on the gel to recover each plasmid in one band for easier comparison of DNA concentration.