Abstract
The mechanism responsible for the accurate partitioning of newly replicated Escherichia coli chromosomes into daughter cells remains a mystery. In this article, we use automated cell cycle imaging to quantitatively analyse the cell cycle dynamics of the origin of replication (oriC) in hundreds of cells. We exploit the natural stochastic fluctuations of the chromosome structure to map both the spatial and temporal dependence of the motional bias segregating the chromosomes. The observed map is most consistent with force generation by an active mechanism, but one that generates much smaller forces than canonical molecular motors, including those driving eukaryotic chromosome segregation.
Introduction
The fitness of all organisms is dependent on the rapid and faithful replication and segregation of the genome to the daughter cells. Although it has long been appreciated that a mitotic spindle drives chromosome segregation in eukaryotic cells, the dominant mechanism exploited by prokaryotic cells is still debated. Active partitioning systems are known to segregate the low-copy-number plasmids (e.g. P1, R1-16 and F) and homologous systems have been found on the chromosomes of Caulobacter crescentus and Bacillus subtilis and a number of other bacteria (1–7). These active systems are believed to have some functional similarity to spindles but often appear to play a surprisingly limited role: for example, the par genes of B. subtilis are not essential. Intriguingly, no homologous system has yet been discovered in Escherichia coli, and a group of nucleoid structural and segregation genes, including mukBEF, seqA and matP, appear to have supplanted both the bacterial structural maintenance of chromosomes (SMC) and partitioning (par) genes in γ-proteobacteria, suggesting that other mechanisms of segregation may play an important role (8,9).
Much of what is known about the E. coli chromosome segregation mechanism is phenomenological and qualitative: In slow growing cells (generation time ∼120 min), the initial locus dynamics is characterized by a Stay-at-Home phenomena where the locus remains localized to mid-cell (10–16). Replication is initiated at the chromosomal origin of replication (oriC), and proceeds bi-directionally down the two arms of the circular chromosome (Figure 1Α) (17). After roughly 20 min of cohesion (18), newly replicated sister loci split and undergo rapid translocation towards the quarter cell positions (the mid-cell location after division). After reaching the quarter cell positions, oriC dynamics is again characterized by a Stay-at-Home phenomenon (11,15). In general the rest of the chromosome is replicated and segregated continuously and sequentially, such that genes sequentially closer to oriC are replicated and segregated earlier than distant genes (13,18). A number of subtle nucleoid structural transitions have also been reported (T1, T2 and T3), in which loci on the right arm of the chromosome split cooperatively (19,20).
In this article, we perform a quantitative analysis of the motion of oriC, one of the first loci to segregate (16,19). By combining time-lapse epi-fluorescence microscopy with high-throughput automated image analysis, we are able to capture oriC dynamics throughout the cell cycle for greater than an order-of-magnitude more cells than have ever been characterized. This collection of complete cell cycle trajectories facilitates the quantitative analysis of the locus motion summarized qualitatively above. We report the following findings: (i) Mean-Squared Displacement (MSD) analysis of the Rapid-Translocation phase of oriC motion shows sub-diffusive dynamics, rather than processive dynamics. (ii) Similar dynamics are observed for the actively partitioned plasmid R1-16 by MSD analysis, demonstrating that processive dynamics on times scales shorter than a cell cycle are not a prerequisite for active segregation mechanisms. (iii) A comparison of the step-size distribution between the Rapid-Translocation and Stay-at-Home phases of locus motion shows a distribution-wide bias towards the eventual destination, rather than the presence of large biased steps. (iv) Faithful segregation of the origin loci results from a small diffusional bias, a drift velocity, that switches from a restoring force, centred around mid-cell before locus segregation, to a restoring force centred around the quarter cell positions immediately proceeding locus splitting. The cell appears to identify the quarter cell positions in advance of the arrival of oriC suggesting the existence of a cellular landmark determining this position. Because the nucleoid is significantly remodelled during this period while the drift velocity remains qualitatively unchanged, it is unlikely that nucleoid structure (19,20) or chromosome entropy (21) is the dominant source of the diffusional bias and therefore suggests the existence of an additional as-yet undiscovered segregation mechanism in E. coli. The measurement of the drift velocity and the interpretation of this velocity in terms of a driving force provide the first clear biophysical picture of the dynamical changes that drive the segregation process and reconcile the seemingly conflicting observations of sub-diffusive MSD scaling and active segregation. We expect this analysis to be applicable not only to the interpretation of other chromosome dynamics problems, but also to sub-cellular stochastic motion in general.
The mechanism responsible for the accurate partitioning of newly replicated Escherichia coli chromosomes into daughter cells remains a mystery. In this article, we use automated cell cycle imaging to quantitatively analyse the cell cycle dynamics of the origin of replication (oriC) in hundreds of cells. We exploit the natural stochastic fluctuations of the chromosome structure to map both the spatial and temporal dependence of the motional bias segregating the chromosomes. The observed map is most consistent with force generation by an active mechanism, but one that generates much smaller forces than canonical molecular motors, including those driving eukaryotic chromosome segregation.
IntroductionThe fitness of all organisms is dependent on the rapid and faithful replication and segregation of the genome to the daughter cells. Although it has long been appreciated that a mitotic spindle drives chromosome segregation in eukaryotic cells, the dominant mechanism exploited by prokaryotic cells is still debated. Active partitioning systems are known to segregate the low-copy-number plasmids (e.g. P1, R1-16 and F) and homologous systems have been found on the chromosomes of Caulobacter crescentus and Bacillus subtilis and a number of other bacteria (1–7). These active systems are believed to have some functional similarity to spindles but often appear to play a surprisingly limited role: for example, the par genes of B. subtilis are not essential. Intriguingly, no homologous system has yet been discovered in Escherichia coli, and a group of nucleoid structural and segregation genes, including mukBEF, seqA and matP, appear to have supplanted both the bacterial structural maintenance of chromosomes (SMC) and partitioning (par) genes in γ-proteobacteria, suggesting that other mechanisms of segregation may play an important role (8,9).
Much of what is known about the E. coli chromosome segregation mechanism is phenomenological and qualitative: In slow growing cells (generation time ∼120 min), the initial locus dynamics is characterized by a Stay-at-Home phenomena where the locus remains localized to mid-cell (10–16). Replication is initiated at the chromosomal origin of replication (oriC), and proceeds bi-directionally down the two arms of the circular chromosome (Figure 1Α) (17). After roughly 20 min of cohesion (18), newly replicated sister loci split and undergo rapid translocation towards the quarter cell positions (the mid-cell location after division). After reaching the quarter cell positions, oriC dynamics is again characterized by a Stay-at-Home phenomenon (11,15). In general the rest of the chromosome is replicated and segregated continuously and sequentially, such that genes sequentially closer to oriC are replicated and segregated earlier than distant genes (13,18). A number of subtle nucleoid structural transitions have also been reported (T1, T2 and T3), in which loci on the right arm of the chromosome split cooperatively (19,20).
In this article, we perform a quantitative analysis of the motion of oriC, one of the first loci to segregate (16,19). By combining time-lapse epi-fluorescence microscopy with high-throughput automated image analysis, we are able to capture oriC dynamics throughout the cell cycle for greater than an order-of-magnitude more cells than have ever been characterized. This collection of complete cell cycle trajectories facilitates the quantitative analysis of the locus motion summarized qualitatively above. We report the following findings: (i) Mean-Squared Displacement (MSD) analysis of the Rapid-Translocation phase of oriC motion shows sub-diffusive dynamics, rather than processive dynamics. (ii) Similar dynamics are observed for the actively partitioned plasmid R1-16 by MSD analysis, demonstrating that processive dynamics on times scales shorter than a cell cycle are not a prerequisite for active segregation mechanisms. (iii) A comparison of the step-size distribution between the Rapid-Translocation and Stay-at-Home phases of locus motion shows a distribution-wide bias towards the eventual destination, rather than the presence of large biased steps. (iv) Faithful segregation of the origin loci results from a small diffusional bias, a drift velocity, that switches from a restoring force, centred around mid-cell before locus segregation, to a restoring force centred around the quarter cell positions immediately proceeding locus splitting. The cell appears to identify the quarter cell positions in advance of the arrival of oriC suggesting the existence of a cellular landmark determining this position. Because the nucleoid is significantly remodelled during this period while the drift velocity remains qualitatively unchanged, it is unlikely that nucleoid structure (19,20) or chromosome entropy (21) is the dominant source of the diffusional bias and therefore suggests the existence of an additional as-yet undiscovered segregation mechanism in E. coli. The measurement of the drift velocity and the interpretation of this velocity in terms of a driving force provide the first clear biophysical picture of the dynamical changes that drive the segregation process and reconcile the seemingly conflicting observations of sub-diffusive MSD scaling and active segregation. We expect this analysis to be applicable not only to the interpretation of other chromosome dynamics problems, but also to sub-cellular stochastic motion in general.
