Figure 9: Dumbbell-Shaped Nucleoids are Intermediates of Chromosome Segregation--Compact nucleoids were purified from C600 and C600<em>parC1215</em> and visualized by fluorescence microscopy after staining with DAPI. Two morphologically distinct classes are present in nucleoids purified from <em>E.coli</em> cells: (A) Singlets and (B) dumbbells.
Figure 10: Filaments of MreB--Electron microscopic analysis of MreB filaments. MreB was diluted to a concentration of either 100 nM (D and E), 1 μM (C, F, and G), or 4 μM (H) and polymerized at 37 °C for either 5 min (C) or 30 min (D-H). Filaments were visualized by electron microscopy. (I) Distribution of the length of the filaments formed at the indicated concentrations of MreB.
Figure 11: The Oligomeric State of MreB Determines its Effect on Topo IV Decatenation Activity--(A) Two sets of reactions were performed. In one set (lanes marked "M"), monomeric MreB was introduced into Topo IV kDNA decatenation reactions. kDNA is mitochondrial DNA obtained from trypanosomes such as <em>Crithidia fasciculata</em> and consists of large networks (~5000) of mostly covalently closed, catenated minicircle. Because of their high molecular weight, these networks are unable to enter agarose gels. However, type II topoisomerases such as Topo IV and can unlink the minicircles that can then be detected by agarose gel electrophoresis. In the second set of reactions (lanes marked "P"), the indicated amounts of MreB were pre-incubated for 30 min at 37 °C to form filaments prior to the addition of Topo IV. (B) Quantification of (A).
Figure 12: Interaction Between MreB and Topo IV May Help Coordinate Late Cell Cycle Events--(1) Topo IV activity is spatially and temporally regulated during the bacterial cell cycle. (2) Active Topo IV is assembled only late in the cell cycle. Meanwhile, remodeling of MreB filaments during the cell cycle causes the accumulation of a ring of polymerized MreB at the cell center. (3) Polymerized MreB stimulates the decatenation activity of Topo IV. (4) MreB filaments are further remodeled to facilitate their segregation to the two daughter cells. This remodeling produces monomeric MreB. (5) Inhibition of Topo IV activity by monomeric MreB restores a state of low Topo IV activity at the onset of a new cell cycle.
Figure 13: The MinCDE System is the Target of the DNA Decondensation Checkpoint--MinD-GFP, which displays a pole-to-pole oscillation in WT cells, does not oscillate in the <em>dnaXE145A</em> strain (this mutation delocalizes Topo IV, leading to promiscuous Topo IV activity, decondensation of the nucleoid, and activation of the cell division checkpoint). (A) are wild-type cells; (B) are filaments formed after treatment of wild-type cells with cephalexcin to inhibit cell division, note that MinD still oscillates in striped zones; (C) are <em>dnaXE145A</em> cells grown in rich media for 3 h. This lack of oscillation can account for the observed division inhibition.
Figure 14: The Interaction Between MukB and ParC is Required for Stimulation of Topo IV-Catalyzed Superhelical DNA Relaxation--(A) MukB stimulates wild-type Topo IV-catalyzed relaxation of negatively supercoiled DNA. However, Topo IV reconstituted from ParE and ParCR705A, a variant ParC protein that cannot interact with MukB, is not stimulated by MukB.
Figure 15: Resolution of Converging, Stalled Replication Forks by RecQ and Topo III--(A) Schematic of substrate preparation and resolution. Plasmid DNA (i) is replicated in the presence of Tus, generating a late replicative intermediate (LRI, iii) where the 3'-OH ends of the nascent leading strands of the converging replication forks are separated by about 130 bp (iv). Treatment of the LRI with RecQ and Topo III yields two gapped, form II sister DNA molecules. (B) Products of the replication reaction in either the absence or presence of Tus. (C) The LRI resolution reaction. Complete resolution of the LRI to daughter form II DNA molecules requires Topo III, RecQ, and SSB.
Our interests in the mechanisms of chromosome segregation arose from our studies on the roles during DNA replication of the four E. coli DNA topoisomerases. In previous studies we have shown that the subunits of topoisomerase IV, the topoisomerase that decatenates the replicated sister chromosomes, are found in different places in the cell: ParC is found at the cell center associated with a number of replisome proteins and ParE is distributed throughout the cell. Furthermore, Topo IV activity manifests only late in the cell cycle, when DNA replication is nearly complete and is focused at the cell center. For the past few years we have been interested in the following questions: Why and how is Topo IV activity regulated? What is the meaning of the physical separation in the cell of the subunits of Topo IV? And, is there an alternative pathway of chromosome decatenation in the cell?
Overexpression of an ATPase-defective, dominant-negative allele of mreB, which encodes the bacterial actin homolog that forms helical filaments in the cell, generates a par phenotype similar to that observed when either parE10 or parC1215 mutants are cultured at the non-permissive temperature. Because of this similarity, we asked whether MreB affected the activity of Topo IV. We found that mreB mutant cells accumulated dumbbell-shaped nucleoids (Figure 9) that were intermediates in chromosome segregation and could be decatenated in vitro by Topo IV. Overexpression of Topo IV in these cells also suppressed the accumulation of the dunbbell-shaped nucleoids. We purified E. coli MreB and showed that it had a differential effect on Topo IV activity that depended on its native state (Figure 10): monomeric MreB stimulated Topo IV whereas polymerized MreB inhibited it (Figure 11). We believe this differential effect can account in part for the temporal regulation of Topo IV activity and also explain why mreB mutants display chromosome segregation defects. Our model (Figure 12) posits that when Topo IV is formed in the cell center at the end of the cell cycle the polymerized MreB stimulates its activity. The chromosomes become unlinked and the MreB helix must be remodeled to allow cytokinesis to occur. This event creates a localized increase in the concentration of monomeric MreB, which then inhibits Topo IV.
Some recent studies suggest an answer to the question of why Topo IV activity is regulated. We have found that promiscuous expression of Topo IV throughout the cell cycle causes nucleoid decondensation that, in turn, causes an inhibition of cell division. This effect appears to be a general phenomenon in that three other conditions that cause the nucleoid to decondense also cause an inhibition of cell division. This inhibition is not a function of induction of the SOS response and defects in chromosome segregation, for example, abrogating Topo IV activity by raising parC1215 cells to the non-permissive temperature, do not trigger the response. This dependency of cell division on the state of the nucleoid suggest a surveillance mechanism that operates as a checkpoint. We have identified the target of the checkpoint as the MinCDE system that, by oscillating from pole to pole, prevents septal ring formation everywhere in the cell but in the center. MinC/D oscillation is inhibited in cell filaments that contain decondensed chromosomes (Figure 13). We are very interested in the mechanism of execution of this checkpoint.
Physical separation of the subunits of Topo IV raise the possibility that they might play another role in the cell in addition to constituting a topoisomerase. Along these lines we have found that ParC exhibits a strong interaction with MukB, the bacterial condensin, both in vivo and in vitro. Furthermore, MukB stimulates Topo IV superhelical DNA relaxation activity. We have mapped the site on ParC that interacts with MukB and generated variant ParC proteins that no longer interact with MukB. Interestingly, these variant ParC proteins still reconstitute active Topo IV with ParE, but this enzyme is no longer stimulated by MukB (Figure 14). Strains carrying the variant alleles of parC exhibit a partially penetrant chromosome segregation phenotype. We are very interested in determining the underlying basis of this phenotype and whether this phenotype is pointing to a MukB- and ParC-mediated effect on chromosome organization.
Overexpression of topoisomerase III, a type IA enzyme, can rescue the partition defect of the parC1215 mutant at the non-permissive temperature, suggesting an alternate pathway of chromosome decatenation. We have reconstituted a reaction that we believe can account for how Topo III can unlink the daughter chromosomes. It involves the resolution of the region between two opposing replication forks by the combined action of a DNA helicase, in this case RecQ, unwinding the unreplicated parental DNA while Topo III unlinks the single-stranded linkages that are generated (Figure 15). This reaction pathway may play a role in the contribution that RecQ and Topo III make to the maintenance of genomic integrity in eukaryotes. Interestingly, we know that RecQ is not required for Topo III suppression of Topo IV defects in vivo, implying that another helicase in the cell may be involved. We are actively trying to identify the cellular participants of this alternate chromosome segregation pathway.