Measuring chromosome dynamics on different time scales using resolvases with varying half-lives

Abstract

The bacterial chromosome is organized into multiple independent domains, each capable of constraining the plectonemic negative supercoil energy introduced by DNA gyrase. Different experimental approaches have estimated the number of domains to be between 40 and 150. The site-specific resolution systems of closely related transposons Tn3 and gammadelta are valuable tools for measuring supercoil diffusion and analysing bacterial chromosome dynamics in vivo. Once made, the wild-type resolvase persists in cells for time periods greater than the cell doubling time. To examine chromosome dynamics over shorter time frames that are more closely tuned to processes like inducible transcription, we constructed a set of resolvases with cellular half-lives ranging from less than 5 min to 30 min. Analysing chromosomes on different time scales shows domain structure to be dynamic. Rather than the 150 domains detected with the Tn3 resolvase, wild-type cells measured over a 10 min time span have more than 400 domains per genome equivalent, and some gyrase mutants exceed 1000.

Fig. 1

Half-life (t_1/2) measurement of modified resolvases. Strains with plasmids expressing different cloned resolvase proteins were grown to mid-log phase (Klett 50) at 30°C, induced at 42°C for 10 min, returned to 30°C and samples were subsequently periodically harvested by centrifugation. After cells were resuspended in 10% ice-cold glycerol, plasmid pRR51 was introduced by electroporation. All experiments were performed in triplicate with at least 300 bacterial colonies examined in each test to determine one standard deviation.

Fig. 2

Western blot analysis of resolvase stability. Bacterial cultures (1.5 ml) in mid-log phase (Klett 50) were harvested by centrifugation and lysed in a final volume of 50 μl. Clarified lysate (15 μl) from each preparation was loaded on the gel. Anti-SsrA antibodies identify a band at the predicted size. A non-specific band of higher molecular weight served as the loading control. A. γδRes-SsrA is a short-lived resolvase. The protein is detected at induction (lane 2) and 2 and 4 min after induction (lanes 3 and 4 respectively). After the 6 min post-induction point (lane 5) the protein is at or below detection limits (lanes 6, 7 and 8). The lanes are: n, non-induced; i, induction; 2 min, 4 min, 6 min, 10 min, 15 min and 30 min are post-induction time points. Sample preparation is described under Experimental procedures. B. In a clpP⁻ background (clpP replaced with a kan gene), γδRes-SsrA becomes a very stable protein, detectable for hours after heat induction. The lanes are: n, non-induced; i, induced; 0.5 h, 1 h, 1.5 h and 2 h are post-induction time points. Sample preparation is described under Experimental procedures.

Fig. 3

A. Resolution assays for nine intervals ranging in size from 12 to 90 kb in the his-cob region. Three independent experiments were averaged for each time point, and at least 300 colonies were counted for each time point. The distance penalty for recombination (the slope of the curve) is much lower for the WT γδ resolvase than for the short-lived, ssrA-tagged resolvase. The half-distance value calculated for the wild-type resolvase was 36 kb, and for the ssrA-tagged resolvase was 9 kb. B. The bacterial chromosome changes over time. To illustrate the change, the first-order decay constants for different resolvases in Table 2 were plotted against the time that each enzyme persists in the cell. The resolvase exposure time is defined as the period it takes each type of resolvase to decay so that the resolution activity is half the maximum amount. Note that this value is not equivalent to the half-life for stable enzymes because saturating levels of resolvase persist for periods longer than the cell division time. The enzymes arranged in increasing exposure levels are γδRes-SsrA (1/2D = 9 kb), γδRes-SsrA-A8D (1/2D = 12 kb), γδRes-SsrA-L9D (1/2D = 13 kb), Tn3 Res (1/2D = 15 kb) and γδRes-SsrA-DD (1/2D = 23 kb). Data for the γδRes protein (1/2D = 36 kb) is not included because it remains at saturating levels for more than 6 h after induction.

Fig. 4

Resolution assays with the γδRes-SsrA-DD protein in the presence of wild-type (WT) gyrase (open circles) gyrA209 (open triangles) and gyrB1820 (open squares). The 1/2D values for recombination calculated for each curve were: 23 kb for the WT strain, 18 kb for a strain with gyrA209 and 12 kb for a strain with the gyrB1820 gene.

Fig. 5

Resolution efficiency of a 14 kb domain in strains carrying different DNA gyrase mutations. A 14 kb interval was examined with resolvases having the indicated half-life. The recombination efficiency of each resolvase is reported for: wild-type gyrase, grey bars; gyrA205, hatched bars; gyrA213, horizontally striped bars; gyrA209, diagonally striped bars, and gyrB1820, black bars.