plasmids have active partitioning systems, systems that have counterparts in the chromosomes of many bacteria (but not E. coli). These partitioning (Par) systems consist of two proteins, ParA and ParB, and a centromere‐like DNA site called parS (Baxter and Funnell 2014; Gerdes et al. 2010). The ParB protein binds to parS and ParA interacts with ParB, hydrolysing ATP or GTP to provide the energy needed to drive the partitioning process.
Plasmid Par systems, such as those in the single‐copy F plasmid or the P1 prophage plasmid, whose ParA protein has a Walker‐type ATPase motif, use the surface of the nucleoid as a scaffold over which plasmids are actively moved. The mechanism is termed a diffusion‐ratchet, with ParA diffusing over the nucleoid and ParB binding to the parS sequence on the plasmid to form the partition complex (Vecchiarelli et al. 2013, 2014). ParA‐ParB interaction triggers ATP hydrolysis by ParA, denuding the nucleoid surface in the vicinity of the plasmid parS‐ParB complex of active ParA. This depletion effect creates a ParA gradient across the nucleoid surface, moving the parS‐ParB complex (and the plasmid) along the gradient. With two daughter plasmids in play, the effect of ParA depletion and the associated gradients is to move the two plasmids away from each other, segregating them into the two daughter cells. This diffusion ratchet mechanism has replaced earlier hypothetical models of ParA‐ParB‐parS segregation systems that were based on ParA assembly into cytoskeletal filaments (Brooks and Hwang 2017).
The R1 drug‐resistance single‐copy plasmid has a ParMRC partitioning system that consists of a centromere‐like parC site, an adaptor protein ParR that binds to parC and an actin‐like ATPase, ParM. ParM forms filaments that grow bidirectionally, with a ParR‐parC complex one either end. As the filament grows in length, the plasmid copies are separated. ParM searches the cell for ParR‐parC complexes, the complexes stabilise ParM filaments whose dynamic instability requires ATP hydrolysis; the stabilised filaments grow, pushing parC‐containing plasmids to opposite ends of the cell (Garner et al. 2004, 2007). The TubZ‐TubR‐tubZ partitioning system found in many plasmids in Bacillus spp. (e.g. B. thuringiensis) differs from ParMRC in that the TubZ filament grows unidirectionally by recycling TubZ subunits from the leading edge to extend the trailing edge (‘treadmilling’) and uses GTP hydrolysis to form the filament (Fink and Löwe 2015; Larsen et al. 2007).
1.16 The Nucleoid
A defining characteristic of prokaryotes is that they do not possess a membrane‐bound nucleus. Instead, prokaryotes have a nucleoid, a body within the cytoplasm that contains the genetic material but lacks a surrounding membrane (Piekarski 1937). The nucleoid is composed of the chromosome and associated molecules including RNA polymerase, DNA polymerase, DNA‐binding proteins, and RNA molecules (Dorman 2014b; Macavin and Adhya 2012). In electron micrographs of thinly sectioned bacteria, the nucleoid can be seen as an amoeboid shape surrounded by the electron‐dense ribosomes within the cytoplasm (Kellenberger 1952; Robinow and Kellenberger 1994). Staining of the DNA with 4′, 6‐diamidino‐2‐phenylindole (DAPI) has confirmed the presence of a zone around the nucleoid in E. coli and B. subtilis where translation can take place (Mascarenhas et al. 2001). While even more sophisticated imaging has improved our knowledge of the structure of the nucleoid, it has taken a multi‐pronged approach using a variety of techniques over several decades to bring about our current (but still incomplete) understanding of the bacterial nucleoid.
The chromosome is packaged within the bacterial cell in a conformation that permits gene expression and DNA replication to proceed. The 4.6 Mb circular chromosome of E. coli strain MG1655 has a circumference of 1.5 mm and, if opened out fully, a diameter of approximately 0.5 mm. In contrast, the bacterial cell is approximately 2 μm in length, 1 μm in diameter and has a volume of 1 fl, or 1 × 10−15 l (Dorman 2013; Kubitschek and Friske 1986). Understanding how the need to package the DNA efficiently is reconciled with the requirements of DNA replication, gene transcription, DNA recombination, and DNA repair is a major goal of research into the structure of the nucleoid.
1.17 The Chromosome Has Looped Domains
Examination of electron micrographs of thinly sectioned E. coli cells gives few clues as to the fine detail of chromosome organisation in the nucleoid. The images of chromosomes extruded from gently lysed E. coli obtained by Cairns (1963a,b) using autoradiography hint at a subdivision of the chromosome into looped, supercoiled domains. The nature of the domain boundaries was obscure but seemed to be associated with RNA (Kavenoff and Bowen 1976). Analysis with electron microscopy suggested that the chromosome was subdivided into between 12 and 80 supercoiled loops (Delius and Worcel 1974).
Sinden and Pettijohn (1981) used photobinding of trimethylpsoralen to intracellular DNA to estimate the number of independently looped domains. This agent binds to duplex DNA at a rate that is proportional to the superhelical tension in the DNA (Sinden et al. 1980). By estimating the number of gamma‐radiation‐induced nicks that were required to release most of the superhelical tension, an estimate of the number of topologically independent domains was obtained. For E. coli growing with a generation time of 30 minutes, it was estimated that the chromosome is divided into between 33 and 53 independent domains (or between 90 and 150 domains per nucleoid) (Sinden and Pettijohn 1981). The existence of independent domains is an important concept in nucleoid architecture: genes in one domain may be isolated from DNA topological changes taking place in other domains, making gene location along the chromosome significant for reasons other than differences in copy number arising as a result of gene distance from oriC.
More refined measurements of domain number in the E. coli chromosome or that of its close relative, Salmonella enterica serovar Typhimurium, have been made by exploiting site‐specific recombination reactions that can take place within but not between domains, by counting the number of looped domains in multiple images of extruded chromosomes, and by releasing superhelical tension from domains using restriction enzymes and observing the distance over which an effect on the transcription of a supercoiling‐sensitive gene can be exerted (Postow et al. 2004; Stein et al. 2005). The results from these experimental approaches indicate that the chromosome is subdivided into about 400 independent domains in E. coli cells during exponential growth, with each being approximately 12–14 kb in length. The boundaries between the domains do not seem to be fixed and fewer of them are found in bacteria that have entered stationary phase (Staczek and Higgins 1998).
1.18 The Macrodomain Structure of the Chromosome
The bacteriophage lambda integrase‐mediated site‐specific recombination system has been exploited in studies of nucleoid organisation in E. coli and Salmonella (Garcia‐Russell et al. 2004; Valens et al. 2004). Recombination between copies of the lambda attachment site requires physical contact between the sites and these can be created by random collision (Crisona et al. 1999; Dorman and Bogue 2016). Sites placed at different distances from one another around the chromosome can be assessed for interaction frequency, providing an estimate of the frequency of contact between different parts of the chromosome. At the same time, regions of the chromosome that rarely interact have also been discovered. This analysis led to the proposal that the chromosome is divided into a small number of large territories called macrodomains (Valens et al. 2004) (Figure 1.1). E. coli and its close relatives have four macrodomains (Ori, Left, Ter, and Right) and two non‐structured (NS) regions: NS‐Left and NS‐Right (Cameron et al. 2017; Jiang et al. 2015; Thiel et al. 2012). The NS domains are determined by their proximity to the Ori macrodomain: any region that is placed next to Ori acquires the features of an NS domain (Duigou and Boccard 2017).
1.19 The Chromosome Displays Spatial Arrangement Within the Cell
The bacterial chromosome is oriented in the cytoplasm with reference to the poles and the midpoint of the cell (
