No abstract

SummaryChromosome dimers, formed by homologous recombination between sister chromosomes, normally require cell division to be resolved into monomers by sitespecific recombination at the dif locus of Escherichia coli. We report here that it is not in fact cell division per se that is required for dimer resolution but the action of the cytoplasmic domain of FtsK, which is a bifunctional protein required both for cell division and for chromosome partition.

No abstract

The dif locus is a RecA-independent resolvase site in the terminus region of the chromosome of Escherichia coli. The locus reduces dimer chromosomes, which result from sister chromatid exchange, to monomers. A density label assay demonstrates that recombination occurs at dif, and that it requires XerC and XerD. The frequency of this recombination is approximately 14% per site per generation, which is doubled in polA12 mutants. We have determined that recombination occurs late in the cell cycle, and that resolution is blocked if cell division is inhibited with cephalexin or by a ftsZts mutation. Fluorescence microscopy has demonstrated that abnormal nucleoids are present in cells incubated in cephalexin, and this is increased in polA12 mutants.

The terminus region of the chromosome of Escherichia coli contains two separate sites, called Ti and T2, that inhibit replication forks. Ti is located near 28.5 min, which is adjacent to tip, and T2 is located at 34.5-35.7 min on the opposite side of the terminus region, near manA. The sites act in a polar fashion, and replication forks traveling in a clockwise direction with respect to the genetic map are not inhibited as they pass through Ti but are inhibited at T2. Similarly, counterclockwise forks are not inhibited at T2 but are inhibited at Ti. Consequently, forks are not inhibited until they have passed through the terminus region and are about to leave it. Studies with deletion strains have located T2 within a 58-kilobase interval, which corresponds to kilobase coordinates 387-445 on the physical map of the terminus region.The terminus region of the chromosome of Escherichia coli is located directly opposite the origin of replication on the circular genetic map (1). One of the most interesting features of this region is that it inhibits replication forks that are traveling in either a clockwise or counterclockwise direction with respect to the map (2-5). When origins of replication located near the terminus region were used (2-5), inhibition of replication forks occurred somewhere in the interval between trp (28 min) and manA (36 min). Although the function of this inhibition is unknown, its presence suggests that the terminus might encode partitioning, decatenation, or cell-division control sites, and impediment of replication forks in this region evolved to ensure efficient use of such sites (6).More recent experiments using replication cycles initiated at oriC (7) indicated that the last DNA to be replicated, and consequently the region where forks meet most frequently, was located near 31.2 min. Based on these results, it generally has been assumed that the replication block (terC) is located at this position (1). As discussed by Bouchd et al. (7), however, other interpretations of the data can be made. Therefore, identification of the region of most frequent fork encounter does not necessarily identify the site ofreplicationfork inhibition.We report here, as do de Massy et al. in an accompanying report in this issue (8), that replication forks are not inhibited at a single site in the middle of the terminus region. Instead, there are two inhibition sites, Ti and T2, which are located at the outer edges of the terminus region. These sites are polar, and they only inhibit replication forks that have passed through the terminus region and are about to leave it. Specifically, clockwise traveling replication forks are inhibited at T2 at 34.5-35.8 min, near manA, and counterclockwise traveling forks are inhibited at Ti at 28.5 min, near trp. MATERIALS AND METHODSBacterial Strains. All replication fork assays were performed with strains PK998 or PK1012 and derivatives containing the indicated deletions. These temperature-sensitive strains are dnaA mutants and contained temperature-sensitive P2 D4 cS...

No abstract

In the processes involved in the termination of replication of the circular chromosome of Escherichia coli, it has been determined that the replication terminus is opposite the replication origin and is located somewhere between the trp and his loci, but the position of the replication terminus has not yet been reported with any greater precision (1-4). It is also not certain if the replication terminus is a definite locus on the chromosome or if it is simply wherever the two replication forks involved in bidirectional replication happen to meet. Various investigators have reported results that suggest that replication occurs in a unidirectional fashion in certain situations (5-7). This suggests that there might not be a locus between the trp and his loci that blocks replication forks. Termination has now been studied with several different chromosomes that replicate bidirectionally, and the results demonstrate that bacteriophage X (8) and simian virus 40 (9) do not possess a specific region of the chromosome that blocks replication forks. The plasmid R6K, however, does have a terminus that blocks replication forks (10). The Bacillus subtilis chromosome might also have a specific terminus (11-13).In order to facilitate the study of termination in E. coli, we have sought conditions in which the normal symmetrical pattern of replication of the chromosome was altered so that one replication fork would reach the terminus region earlier than the other. We have recently determined that induction of prophage P2sig5 causes bacterial chromosome replication from the site of insertion of the prophage, and one of the strains we have studied has the prophage integrated near the terminus region. The results of studies of this strain are reported here. MATERIALS AND METHODSBacterial Strains and Media. PK241 is an F-thr leu his arg thi thyA drm dnaA malA+ E. coli strain derived from CRT4624. This latter strain and bacteriophage P2tsD4csigj (which will be called P2sig5 in this paper) were obtained from Y. Hirota (14). PK289 is an F-ilv arg met thi his thyA drm E. coli strain. PM 14 is a met trp his thr thy nic ilv Proteus mirabilis strain. The PM14 F' merogenote strains containing the F'129, F'116, F'111, F'101, and F'152 episomes were obtained by mating PM14 with E. coli strains containing these episomes and selecting for His+, Thy+, Ilv+, Thr+, and Gal+ cells, respectively. The E. coli F' strains were obtained from B. Bachman.The sources of the A specialized transducing phage were as follows: Xgt-E. coli EcoRI lop-li lig+ 1 (Alig), from R. Davis (15); XflaN+ 36 (Xfla), from M. Simon; A reverse (Arev), from M. Gellert (16); XptrpED 1 (Xtrp), from C. Yanofsky (17); XCIam34bio256 (Xbio), from M. Furth. XaroD was obtained by using the procedure of Schrenk and Weisberg (18).Cells were usually grown in M9 medium (19). During incubation in the absence of required amino acids, arginine or methionine was not removed from the medium. When PK241-P2-1 cells were labeled with 32P, they were grown in low-phosphate medium (20) containing 0.2...

‘Newborn’ Escherichia coli B/r cells, obtained by membrane elution, were used to study the cell cycles of wild‐type and Dam methyltransferase mutants. In wild‐type cells, initiation of chromosome replication was synchronous and tightly controlled. In dam mutants, initiation was altered, but not random. We propose that this is due to the absence of an initiation cascade caused by liberated DnaA molecules, and that this cascade normally synchronizes initiation. The dam‐ cells contained mainly two, three or four replication origins, and this affected nucleoid partitioning as well as cell division. In cultures growing with a 50 min doubling time, a variety of cell cycles were present and half the origins were used every 25 min. Some cells had a 25 min interdivision time, whereas others had an interdivision time longer than the generation time. Partitioning of nucleoids containing unequal numbers of replication origins could also be readily observed by fluorescence microscopy in the dam mutant. Based upon these observations we propose that the dam mutant is also an initiation cascade mutant.