Membrane association of active plasmid partitioning protein A in Escherichia coli

Catching a Walker in the Act-DNA Partitioning by ParA Family of Proteins

Partitioning the replicated genetic material is a crucial process in the cell cycle program of any life form. In bacteria, many plasmids utilize cytoskeletal proteins that include ParM and TubZ, the ancestors of the eukaryotic actin and tubulin, respectively, to segregate the plasmids into the daughter cells. Another distinct class of cytoskeletal proteins, known as the Walker A type Cytoskeletal ATPases (WACA), is unique to Bacteria and Archaea. ParA, a WACA family protein, is involved in DNA partitioning and is more widespread. A centromere-like sequence parS, in the DNA is bound by ParB, an adaptor protein with CTPase activity to form the segregation complex. The ParA ATPase, interacts with the segregation complex and partitions the DNA into the daughter cells. Furthermore, the Walker A motif-containing ParA superfamily of proteins is associated with a diverse set of functions ranging from DNA segregation to cell division, cell polarity, chemotaxis cluster assembly, cellulose biosynthesis and carboxysome maintenance. Unifying principles underlying the varied range of cellular roles in which the ParA superfamily of proteins function are outlined. Here, we provide an overview of the recent findings on the structure and function of the ParB adaptor protein and review the current models and mechanisms by which the ParA family of proteins function in the partitioning of the replicated DNA into the newly born daughter cells.

The Coxiella burnetii QpH1 plasmid is a virulence factor for colonizing bone marrow-derived murine macrophages

Coxiella burnetii strains carry one of four large, conserved, autonomously replicating plasmids (QpH1, QpRS, QpDV, and QpDG) or a QpRS-like chromosomally integrated sequence of unknown function. Here we report the characterization of the QpH1 plasmid of C. burnetii Nine Mile phase II by making QpH1-deficient strains. A shuttle vector pQGK containing the CBUA0036-0039a region (predicted as being required for the QpH1 maintenance) was constructed. The pQGK vector can be stably transformed into the Nine Mile II and maintained at a similar low copy like QpH1. Importantly, transformation with pQGK cured the endogenous QpH1 due to plasmid incompatibility. Compared to a Nine Mile II transformant of a RSF1010-ori based vector, the pQGK transformant shows a similar growth curve in both axenic media and Buffalo green monkey kidney cells, a variable growth defect in macrophage-like THP-1 cells depending on the origin of inoculum, and dramatically reduced ability of colonizing wild-type bone marrow-derived murine macrophages. Furthermore, we found CBUA0037-0039 ORFs are essential for plasmid maintenance, and CBUA0037-0038 ORFs account for plasmid compatibility. And plasmid-deficient C. burnetii can be isolated by using CBUA0037 or -0038 deletion vectors. Furthermore, QpH1-deficient C. burnetii strains caused a lesser extent of splenomegaly in SCID mice but, intriguingly, they had significant growth in SCID mouse-sourced macrophages. Taken together, our data suggest that QpH1 encodes factor(s) essential for colonizing murine, not human, macrophages. This study suggests a critical role of QpH1 for C. burnetii persistence in rodents and expands the toolkit for the genetic studies in C. burnetii Author summary All C. burnetii isolates carry one of four large, conserved, autonomously replicating plasmids or a plasmid-like chromosomally integrated sequence. The plasmid is a candidate virulence factor of unknown function. Here we describe the construction of novel shuttle vectors that allow making plasmid-deficient C. burnetii mutants. With this plasmid-curing approach, we characterized the role of the QpH1 plasmid in in vitro and in vivo C. burnetii infection models. We found that the plasmid plays a critical role for C. burnetii growth in murine macrophages. Our work suggests an essential role of the QpH1 plasmid for the acquisition of colonizing capability in rodents by C. burnetii This study represents a major step toward unravelling the mystery of the C. burnetii cryptic plasmids.

Identification of a Potential Membrane-Targeting Sequence in the C-Terminus of the F Plasmid Segregation Protein SopA

Stable maintenance and partitioning of the 'Fertility' plasmid or the F plasmid in its host Escherichia coli require the function of a ParA superfamily of proteins known as SopA. The mechanism by which SopA mediates plasmid segregation is well studied. SopA is a nucleoid-binding protein and binds DNA in an ATP-dependent but sequence non-specific manner. ATP hydrolysis stimulated by the binding of the SopBC complex mediates the release of SopA from the nucleoid. Cycles of ATP-binding and hydrolysis generate an ATPase gradient that moves the plasmid through a chemophoresis force. Nucleoid binding of SopA thus assumes a central role in its plasmid-partitioning function. However, earlier work also suggests that the F plasmid can be partitioned into anucleate cells, thus implicating nucleoid independent partitioning. Interestingly, SopA is also reported to be associated with the inner membrane of the bacteria. Here, we report the identification of a possible membrane-targeting sequence, a predicted amphipathic helix, at the C-terminus of SopA. Molecular dynamics simulations indicate that the predicted amphipathic helical motif of SopA has weak affinity for membranes. Moreover, we experimentally show that SopA can associate with bacterial membranes, is detectable in the membrane fractions of bacterial lysates, and is sensitive to the membrane potential. Further, unlike the wild-type SopA, a deletion of the C-terminal 29 amino acids results in the loss of F plasmids from bacterial cells.

Chromosome loss from par mutants of Pseudomonas putida depends on growth medium and phase of growth

The proteins encoded by chromosomal homologues of the parA and parB genes of many bacterial plasmids have been implicated in chromosome partitioning. Unlike their plasmid counterparts, mutant phenotypes produced by deleting these genes have so far been elusive or weakly expressed, except during sporulation. Here the properties of Pseudomonas putida strains with mutations in parA and parB are described. These mutants do not give rise to elevated levels of anucleate bacteria when grown in rich medium under standard conditions. However, in M9-minimal medium different parA and parB mutations gave between 5 and 10% anucleate cells during the transition from exponential phase to stationary phase. Comparison of the DNA content of bacteria at different stages of the growth curve, in batch culture in L-broth and in M9-minimal medium, suggests that the par genes are particularly important for chromosome partitioning when cell division reduces the chromosome copy number per cell from two to one. This transition occurs in P. putida during the entry into stationary phase in M9-minimal medium, but not in L-broth. It is proposed that the partition apparatus is important to ensure proper chromosome segregation primarily when the bacteria are undergoing cell division in the absence of ongoing DNA replication.

Role of the ATP-binding site of SopA protein in partition of the F plasmid

SopA belongs to a large family of bacterial partition protein ATPases. It helps stabilize the F plasmid by acting as the primary repressor of transcription of the sopAB operon, preventing the destabilizing effects of Sop protein excess. It is also thought to act directly in the F partition mechanism. We have examined the role of SopA in partition and repression by observing the consequences of replacing an invariant ATP-binding site lysine, K120, by glutamine or arginine. Circular dichroism studies of the purified mutant proteins revealed no major differences from wild-type, but in the presence of ADP or ATP each protein showed a characteristic spectrum which suggested a distinct conformational change. The K120Q mutant retained most of the wild-type ATPase activity while the K120R mutant lost it. In neither case was the residual activity stimulated by SopB, as occurs for wild-type SopA. The strength of sop promoter repression by the mutant SopA proteins alone was comparable to that resulting from SopB-enhancement of wild-type SopA, but SopB enhanced repression by the mutant SopA proteins either slightly (K120R) or not at all (K120Q). Mini-Fs in which the sop operon was controlled by a constitutive promoter were destabilized by the mutations, demonstrating the need for SopA and its ATP-binding site in the partition process. The K120R mini-F was lost at the same rate as a mini-F lacking the sopC centromere, the K120Q mutant was lost faster. SopAK120R at high levels was more effective than SopA(+) in disrupting the partition complex, whereas SopAK120Q did not disrupt it at all. These results suggest that one function of SopA in the partition mechanism is to break the paired plasmid structure to allow F molecules to segregate to daughter cells.

The bacterial ParA-ParB partitioning proteins

A pair of genes designated parA and parB are encoded by many low copy number plasmids and bacterial chromosomes. They work with one or more cis-acting sites termed centromere-like sequences to ensure better than random predivisional partitioning of the DNA molecule that encodes them. The centromere-like sequences nucleate binding of ParB and titrate sufficient protein to create foci, which are easily visible by immuno-fluorescence microscopy. These foci normally follow the plasmid or the chromosomal replication oriC complexes. ParA is a membrane-associated ATPase that is essential for this symmetric movement of the ParB foci. In Bacillus subtilis ParA oscillates from end to end of the cell as does MinD of E. coli, a relative of the ParA family. ParA may facilitate ParB movement along the inner surface of the cytoplasmic membrane to encounter and become tethered to the next replication zone. The ATP-bound form of ParA appears to adopt the conformation needed to drive partition. Hydrolysis to create ParA-ADP or free ParA appears to favour a form that is not located at the pole and binds to DNA rather than the partition complex. Definition of the protein domains needed for interaction with membranes and the conformational changes that occur on interaction with ATP/ADP will provide insights into the partitioning mechanism and possible targets for inhibitors of partitioning.

Dynamic localization of bacterial and plasmid chromosomes

Plasmid-encoded partition genes determine the dynamic localization of plasmid molecules from the mid-cell position to the 1/4 and 3/4 positions. Similarly, bacterial homologs of the plasmid genes participate in controlling the bidirectional migration of the replication origin (oriC) regions during sporulation and vegetative growth in Bacillus subtilis, but not in Escherichia coli. In E. coli, but not B. subtilis, the chromosomal DNA is fully methylated by DNA adenine methyltransferase. The E. coli SeqA protein, which binds preferentially to hemimethylated nascent DNA strands, exists as discrete foci in vivo. A single SeqA focus, which is a SeqA-hemimethylated DNA cluster, splits into two foci that then abruptly migrate bidirectionally to the 1/4 and 3/4 positions during replication. Replicated oriC copies are linked to each other for a substantial period of generation time, before separating from each other and migrating in opposite directions. The MukFEB complex of E. coli and Smc of B. subtilis appear to participate in the reorganization of bacterial sister chromosomes.

Disruption of the F plasmid partition complex in vivo by partition protein SopA

The SopA protein plays an essential, though so far undefined, role in partition of the mini-F plasmid but, when overproduced, it causes loss of mini-F from growing cells. Our investigation of this phenomenon has revealed that excess SopA protein reduces the linking number of mini-F. It appears to do so by disturbing the partition complex, in which SopB normally introduces local positive supercoiling upon binding to the sopC centromere, as it occurs only in plasmids carrying sopC and in the presence of SopB protein. SopA-induced reduction in linking number is not associated with altered sop promoter activity or levels of SopB protein and occurs in the absence of changes in overall supercoil density. SopA protein mutated in the ATPase nucleotide-binding site (K120Q) or lacking the presumed SopB interaction domain does not induce the reduction in linking number, suggesting that excess SopA disrupts the partition complex by interacting with SopB to remove positive supercoils in an ATP-dependent manner. Destabilization of mini-F also depends on sopC and SopB, but the K120Q mutant retains some capacity for destabilizing mini-F. SopA-induced destabilization thus appears to be complex and may involve more than one SopA activity. The results are interpreted in terms of a regulatory role for SopA in the oligomerization of SopB dimers bound to the centromere.

Sequencing and characterization of the cryptic plasmid QpRS from Coxiella burnetii

Plasmid QpRS from Coxiella burnetii, isolate Priscilla Q177, phase I, has been completely sequenced. DNA for sequencing was amplified with "extra-long" (XL) PCR. The size of the QpRS plasmid sequence was determined to be 39,280 bp, with a G + C content of 39.3%. Putative proteins associated with replication and recombination were identified. The sequence of QpRS plasmid was analyzed for shared and unique regions among C. burnetii plasmids.

The F plasmid centromere, sopC, is required for full repression of the sopAB operon

The SopB protein of the F plasmid has a dual role in the partition of F plasmid copies to daughter cells prior to division. It binds to the sopC centromere site to form the partition complex needed for stabilizing the plasmid, and it interacts with SopA to repress transcription of the sopAB operon, thus preventing the destabilization that results from excess SopB. We have isolated sop mutants by screening for unstable inheritance of mutagenized mini-F DNA. Four of the mutants resulted from different missense mutations in sopB. All four were deficient, to varying degrees, in autoregulation of Sop protein synthesis. The mutant proteins showed diminished capacity for reducing the linking number of mini-F and for destabilizing a plasmid carrying sopC, indicating that reduced ability to form a normal complex with sopC might underlie the autoregulation defect. Repression of the transcription of a sop promoter- lacZ fusion by SopA and SopB was strongly enhanced in the presence of sopC, in cis or in trans, and the enhancement was reduced or nullified when wild-type sopB was replaced by the mutant sopB alleles. A single 43 bp unit of sopC was almost as effective as sopC itself in enhancing repression. The results show that sopC is necessary for full repression of the sop promoter. They thus indicate a previously unsuspected role for this centromere site, and suggest that autoregulation and partition might normally be coordinated processes.

P1 ParA interacts with the P1 partition complex at parS and an ATP-ADP switch controls ParA activities

The partition system of P1 plasmids is composed of two proteins, ParA and ParB, and a cis-acting site parS. parS is wrapped around ParB and Escherichia coli IHF protein in a higher order nucleoprotein complex called the partition complex. ParA is an ATPase that autoregulates the expression of the par operon and has an essential but unknown function in the partition process. In this study we demonstrate a direct interaction between ParA and the P1 partition complex. The interaction was strictly dependent on ParB and ATP. The consequence of this interaction depended on the ParB concentration. At high ParB levels, ParA was recruited to the partition complex via a ParA-ParB interaction, but at low ParB levels, ParA removed or disassembled ParB from the partition complex. ADP could not support these interactions, but could promote the site-specific DNA binding activity of ParA to parOP, the operator of the par operon. Conversely, ATP could not support a stable interaction of ParA with parOP in this assay. Our data suggest that ParA-ADP is the repressor of the par operon, and ParA-ATP, by interacting with the partition complex, plays a direct role in partition. Therefore, one role of adenine nucleotide binding and hydrolysis by ParA is that of a molecular switch controlling entry into two separate pathways in which ParA plays different roles.