Cell cycle-coordinated maintenance of the Vibrio bipartite genome

Abstract

To preserve the integrity of their genome, bacteria rely on several genome maintenance mechanisms that are co-ordinated with the cell cycle. All members of the Vibrio family have a bipartite genome consisting of a primary chromosome (Chr1) homologous to the single chromosome of other bacteria such as Escherichia coli and a secondary chromosome (Chr2) acquired by a common ancestor as a plasmid. In this review, we present our current understanding of genome maintenance in Vibrio cholerae, which is the best-studied model for bacteria with multi-partite genomes. After a brief overview on the diversity of Vibrio genomic architecture, we describe the specific, common, and co-ordinated mechanisms that control the replication and segregation of the two chromosomes of V. cholerae. Particular attention is given to the unique checkpoint mechanism that synchronizes Chr1 and Chr2 replication.

Keywords: DnaA; RctB; Vibrio; chromosome; initiation; iteron; multi-partite; plasmid; replication; segregation.

Fig 1

Genome organization of Vibrionaceae reference genomes. The reference strain of E. coli K-12 and 47 Vibrionaceae reference genomes were included in the tree. The concatenated sequences of eight genes (ftsK, gapA, gyrB, mreB, pyrH, recA, rpoA, and topA) were used for phylogenetic analysis as described by Jiang et al. (19). Alignment was performed using MAFFT software (20). Phylogenetic tree was made with IQ-TREE multi-core v.2.0.3 (21). A total of 286 DNA models were tested, and GTR + F + R6 was selected as the best-fitting model to the IQ-TREE-m TESTNEW algorithm according to the Bayesian information criterion. The statistical data were derived from 1,000 ultrafast bootstrap values and 1,000 replicates of SH approximate likelihood ratio test (22). The raw data used to perform this analysis are in Table S1.

Fig 2

Replication regulated sequences are highly similar between V. cholerae and E. coli. (A) Sequence alignment between oriC of E. coli (GenBank: U00096.3) and ori1 from V. cholerae (GenBank: CP028827.1) reveals a highly similar organization. The oriC consists of a DNA unwinding element containing single-stranded binding sites for DnaA (L, M, and R elements and DnaA trios), as well as a DnaA assembly region (DAR) containing double-stranded binding sites for DnaA with strong boxes (R1, R2, and R4) and weak boxes (t1, R5, t2, I2, R3, and I3). The IHF and Fis-binding sequences are highlighted in green and yellow, respectively. GATC Dam-methylation sites are in bold. (B) Alignment of DARS2 site from V. cholerae and E. coli. Strong DnaA boxes (I, II, III, and V) are highlighted in dark blue, and weak DnaA boxes (V-c, V-b, and V-a) are in light blue. Fis-binding site is indicated in yellow. Alignments and annotations were retrieved from references (34 – 37).

Fig 3

Schematic of the origin organization of iteron plasmids and Chr2 of V. cholerae. Initiator genes are displayed in purple and initiator binding sites (iterons and inverted repeats) are displayed in green. The 29m/39m RctB-binding sites (specific to Chr2) are shown in red. DnaA-binding sites are shown in light blue. The incompatibility regions (inc) and DNA unwinding elements (DUE) are indicated. These organization schemes are in scale. The replicon sequences were retrieved from the NCBI database (GenBank: pSC101: X01654.1, pPS10: X58896.1, R6K: LT827129.1, F: AP001918.1, RK2: CP116242.1, P1: OP279344.1, Chr2: CP028828.1).

Fig 4

Domain II and III from RctB are structural homologs to WH1 and WH2 from Rep proteins. (A) Pi_R6K , (B) RepA_pPS10 , (C) RepE_F , and (D) RctB_Chr2. Rep proteins dimerize through their WH1 domain; RctB dimerize through its domain II. RepE and RctB dimer structures were retrieved from PDB (2Z9O and 5UBF, respectively) (58, 63). Pi_R6K and RepA_pPS10 dimer structures have been predicted using ColabFold v.1.5.2 (64) (Fig. S1). Initiator sequences were retrieved from replicon sequences present on the NCBI database (pPS10: X58896.1 and R6K: LT827129.1)

Fig 5

The RctB initiator is organized into four structural domains. (A) RctB is a 658-residue protein that contains four structural domains: I, II, III, and IV. Domains I, II, and III are all involved in DNA binding via an HTH motif. Domains II and III are structurally similar to the WH1 and WH2 domains of iteron plasmid initiators. Domain IV allows oligomerization and is dedicated to regulating Chr2 initiation. (B) Predicted structure of RctB using ColabFold v.1.5.2 (64) (Fig. S1). (C) Structure of RctB with amino acids involved in DNA binding in dark red (58). (D) Schematic depicting the binding of RctB to the iteron array (six iterons) of the V. cholerae Chr2 origin of replication. Note that the precise conformation of RctB to its various sites remains an open question. The same color code as in panel A is used throughout the figure.

Fig 6

Model for replication origin opening in Chr1 and Chr2. (A) ori1 : the origin of replication of Chr1 is a 245-bp-long sequence. It contains three types of predicted DnaA-binding sites (34, 37): strong double-stranded sites (dark blue: R1, R2, and R4); weak double-stranded sites (light blue: T1, R5, T2, I1, I2, C3, R3, C2, I3, and C1); single-stranded sites inside the DNA unwinding element (gray: DUE; red: L, M, and R elements; orange: DnaA trios). ori2min : The minimal Chr2 origin of replication is 258 bp long and contains a strong DnaA box, six regularly interspaced repeats of 12 mer (iterons), an IHF-binding site (IBS), and six tetranucleotide ATCA repeats inside the DUE (67, 68). (B) DnaA and RctB loopback models adapted from references (68, 69, 73). DnaA and RctB initiators are shown in yellow and purple, respectively. For clarity, the oligomerization domains of DnaA (domain I) and RctB (domain IV) are not shown. Association of DnaA with ATP is mandatory for binding to weak DnaA boxes (dark green circles). For binding to strong DnaA boxes, DnaA can be associated with either ATP or ADP (light green circles). IHF is shown in pink; DUE open ssDNA is shown in gray. DnaA-binding sites are displayed in the same color code as panel A.

Fig 7

Chr2 initiation is activated by crtS replication, the timing of which can be modulated by the distance of crtS from the Chr1 origin. (Left panel) Schematic representing the replicative state of Chr1 (red) when Chr2 (green) initiation is triggered. The blue rectangle indicates the location of crtS when replicated. (Middle panel) Schematic representation of marker frequency analysis of exponentially growing cultures with log2 of the number of reads plotted against their relative position on Chr1 (red) and Chr2 (green). The blue dashed lines indicate the number of reads of loci where crtS is located (crtS _VC23, crtS _WT, crtS _VC963, and crtS _VC2238). The positions of ori1 and ori2 are set to 0 for a better visualization of the bidirectional replication. The difference between the number of reads at ter1 and ter2 is shaded in gray and indicated by a double arrow. Figure adapted and based on real sequencing data from published work (15). (Right panel) Effect of crtS relocation on Chr1 and Chr2 termination synchrony.

Fig 8

Model for coordination of replication between Chr1 and Chr2 by crtS. The cooperative binding of RctB to the fully methylated iteron array causes DNA opening and RctB oligomerization on the ssDNA DUE (green arrow). RctB binding to the 29m/39m sites inhibits replication initiation at ori2 by intra-molecular handcuffing and represses its own transcription at 29 (red arrows). Upon replication, crtS triggers ori2 initiation by counteracting the inhibitory effect of the 29m/39m sites (blue arrows). The four domains of RctB are represented (I–IV). RctB dimers (domain II interface) are monomerized by the DnaK/J chaperones. RctB monomers cooperatively bind to the iteron array. RctB binding to 29m/39m requires domain IV-mediated interactions. Lrp enhances RctB binding to crtS. RctB-binding sites are indicated: iteron array (green), 29m and 3m (red), and crtS (blue). The iterons present in the inc region have been omitted for clarity. Positive and negative regulations are represented by arrows associated with (+) and (−), respectively.

Fig 9

V. cholerae chromosome segregation and divisome assembly. The single flagellum of V. cholerae is shown to be localized at the old pole, while Chr1 and Chr2 are arranged longitudinally with their ter regions indicated by broken lines. 3C data are represented by dashed arcs, showing interactions between the left and right replichores of Chr2. Inter-chromosomal contacts are also shown, with ori2 in proximity to crtS, as well as the ter1 and ter2 regions. The partitioning systems (ParAB-parS) of each chromosome are represented by colored halos around the replication origins. (I) In nascent cells, HubP is present at the old pole along with ori1 and the ParAB1-parS1 partition complex. All cell division proteins (early and late) localize to the new pole along with ter1. (II) As the cell cycle progresses to ~50%, cells elongate, and early cell division proteins move to midcell, forming a pre-divisional Z-ring. Late cell division proteins remain at the new pole where HubP starts accumulating. A duplicated copy of ori1 complexed with ParAB1 is recruited by HubP to the new pole. The duplication of crtS has triggered ori2 replication. The duplicated copies of ori2 moved to the 1/4–3/4 positions of the cell. (III) At ~80% of the cell cycle, all division proteins have migrated to midcell, assembling into a tighter Z-ring. The ter2 regions are already segregated, while the ter1 regions are still at the future division site. (IV) At 90% of the cell cycle, constriction begins and HubP is evenly distributed at both poles. ter1 regions are segregated by FtsK anchored to the closing septum.