Data from: Threshold effect of growth rate on population variability of Escherichia coli cell lengths

A long-standing question in biology is the effect of growth on cell size. Here, we estimate the effect of Escherichia coli growth rate (r) on population cell size distributions by estimating the coefficient of variation of cell lengths (CVL) from image analysis of fixed cells in DIC microscopy. We find that the CVL is constant at growth rates less than one division per hour, whereas above this threshold, CVL increases with an increase in the growth rate. We hypothesize that stochastic inhibition of cell division owing to replication stalling by a RecA-dependent mechanism, combined with the growth rate threshold of multi-fork replication (according to Cooper and Helmstetter), could form the basis of such a threshold effect. We proceed to test our hypothesis by increasing the frequency of stochastic stalling of replication forks with hydroxyurea (HU) treatment and find that cell length variability increases only when the growth rate exceeds this threshold. The population effect is also reproduced in single-cell studies using agar-pad cultures and 'mother machine'-based experiments to achieve synchrony. To test the role of RecA, critical for the repair of stalled replication forks, we examine the CVL of E. coli ΔrecA cells. We find cell length variability in the mutant to be greater than wild-type, a phenotype that is rescued by plasmid-based RecA expression. Additionally, we find that RecA-GFP protein recruitment to nucleoids is more frequent at growth rates exceeding the growth rate threshold and is further enhanced on HU treatment. Thus, we find growth rates greater than a threshold result in increased E. coli cell lengths in the population, and this effect is, at least in part, mediated by RecA recruitment to the nucleoid and stochastic inhibition of division.Program to analyze E. coli cell lengths from DIC imagesDIC cell length detection code developed in MATLAB and released into the Open Source community. This code is suited for all elongated cell types and has been optimized for the purposes of analyzing E. coli cells. All the data analyzed in the related manuscript for cell-lengths has been generated using this program.E. coli population cell lengths: mutants, drug-treatment and growth ratesThe file includes the raw-data used to calculate the coefficient of variation of E. coli cell lengths. The data encompasses (as marked in the tabs) the growth rate modulation, mother-mother machine analysis, microc-colony analysis, multiple Hydroxyurea concentrations, Trimethoprim treatment and mutants.20170119-cellLengths-athale-gangan.xlsxVideo legendDescription of the Supplementary Videos S1 through S4.videoLegend-gangan.pdfSupplementary video S1The growth and division of a population of E. coli MG1655 cells on an agar-pad is followed in DIC (green) simultaneously with the HupA-GFP (red). The time-stamp is in units hh:mm. The arrow indicates a cell elongation event corresponding to a delay in nucleoid segregation. The scale bar is 10 μm.videoS1-K12-DIC-hupAGfp.movSupplementary video S2Time-lapse images of continuous culture of E. coli MG1655 cells grown in LB were acquired in DIC microscopy. The channels restricting the movement of the cells correspond to the dead-end channels of the 'mother machine', while the central trench corresponds to the nutrient channel. The interval between each frame is 100 seconds. The time stamp is hh:mm:ss. Scale bar: 5 μm.VideoS2-LBmothermachine.movSupplementary video S3Time-lapse images of E. coli MG1655 cells in continuous culture grown in the 'mother machine' with M9 + succinate as the medium at 37°C were recorded in DIC. Images were acquired every 120 s. The time stamp is hh:mm:ss. Scale bar: 5 μm.VideoS3-M9mothermachine.movSupplementary video S4DIC time series of E. coli MG1655 micro-colony originated from single mother cell and growing on agar pad at 37°C was imaged at an interval of 2 minutes between every frame. The time stamp is hh:mm:ss (Scale bar: 5 μm).VideoS4-DICmicrocolony.mo