Figure 1.
Making and repairing a site-specific DNA double-strand break in the E. coli chromosome.
(A) SbcCD-mediated cleavage of a 246 bp interrupted palindrome inserted into the chromosomal lacZ gene. During replication, the palindrome becomes transiently single-stranded on the lagging-strand template. This allows it to form a DNA hairpin that is cleaved by SbcCD, generating a two-ended DSB. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. The palindrome is highlighted by green arrows. (B) RecBCD-mediated HR. The ends of the break are processed by RecBCD to generate 3′ ssDNA overhangs coated in RecA. RecA searches the genome for a homologous DNA sequence and catalyses strand-invasion. This forms a D-loop and HJs. The D-loop is acted upon by the replisome assembly factor, PriA, which initiates DNA synthesis. The HJs can be acted upon by RuvABC, branch-migrated and resolved. This generates two converging replication forks, which, upon convergence, terminate the repair process. (C) Map of the lacZ region of the E. coli chromosome illustrating the position and sequence of two 3x χ arrays that have been inserted 1.5 Kb either side of the palindrome in order to stimulate recombination in close proximity of the DSB. The 8 bp χ recognition sequence, highlighted in red, is repeated three times. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. Pal represents the position of the palindrome.
Figure 2.
Viability of strains containing the palindrome, grown in 0.2% arabinose.
(A) Chronic exposure to DSBs. Serial dilutions of strains were spotted on LB-agar plates supplemented with either 0.2% arabinose or 0.5% glucose and incubated overnight at 37°C. (B) Acute exposure to DSBs. Serial dilutions of strains containing the palindrome and grown in 0.2% arabinose for either 0, 30, 60, or 90 minutes were spotted on LB-agar plates supplemented with 0.5% glucose and incubated overnight at 37°C. Strains used; Rec+ Pal+ (DL2006), Rec+ Pal− (DL2573), ΔrecA Pal+ (DL2075), ΔrecA Pal− (DL2605), ΔruvAB Pal+ (DL2801), ΔruvAB Pal− (DL2800), ΔrecG Pal+ (DL2511), ΔrecG Pal− (DL2610), ΔruvAB ΔrecG Pal+ (DL4464), ΔruvAB ΔrecG Pal− (DL4465).
Figure 3.
Intermediates of DSBR by 2D agarose gel electrophoresis.
(A) Schematic representation of a 2D gel illustrating the expected positions of 3-way (blue) and 4-way (red) DNA junction migration. (B) Map of the region of the chromosome showing the relative positions of the palindrome (green), the 3x χ arrays (red arrows), endogenous χ sites (black arrows) and the chromosomal coordinates of the relevant MfeI and SacI restriction sites used to generate the proximal, central and distal fragments. The relative position of probes used is indicated by black rectangles. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. (CI) 2D gels of the proximal, central and distal fragments for Rec+ (DL4184), ΔruvAB (DL4243), ΔrecG (DL4311), and ΔruvAB ΔrecG (DL4260) strains containing the palindrome, exposed to 0.2% arabinose for 60 minutes. 3-way and 4-way DNA junctions are highlighted by a blue and red arrow, respectively. (CII) Quantifications (represented as mean ± SEM where n = 3) of total amount of intermediates (3-way plus 4-way DNA junctions) accumulated by 2D gel electrophoresis. (CIII) Quantifications (represented as mean ± SEM where n = 3) of 3-way and 4-way DNA junctions. Statistical analysis was carried out using an unpaired T-test. * represents p<0.05, ** represents p<0.01 and *** represents p<0.005.
Figure 4.
(A) Map of the chromosome showing the three SalI fragments around the DSB. The coordinates of the restriction sites are shown in blue. The palindrome is shown as a green triangle and the 1.5 kb 3x χ arrays are shown as red lines. The relative position of probes are represented by small black rectangles. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. (B–E) PFGs for Rec+(DL4184 and DL4201), ΔruvAB (DL4243 and DL4257), ΔrecG (DL4311 and DL4312) and ΔruvAB ΔrecG (DL4260 and DL4313) strains, respectively. Quantifications are represented as mean ± SEM where n = 3. For each probe, Lane 1 contains DNA isolated from a strain not containing the palindrome, grown for 60 minutes in arabinose (pal− SbcCD+ T60). Lane 2 contains DNA from a strain containing the palindrome, grown for 60 minutes in glucose (pal+ SbcCD− T60). Lane 3 contains DNA from a strain containing the palindrome, prior to the addition of either glucose or arabinose (pal+ SbcCD− T0). Lane 4 contains DNA from a strain containing the palindrome, grown for 60 minutes in arabinose (pal+ SbcCD+ T60). ‘Branched’ indicates signal from the well, ‘linear’ indicates signal from the gel. Quantifications are represented as mean ± SEM where n = 3. Statistical analysis was carried out using a paired T-test. * represents p<0.05, ** represents p<0.01 and *** represents p<0.005.
Figure 5.
2D agarose gel electrophoresis 30 Kb upstream of the DSB.
(A) SalI map of the region surrounding the DSB showing the location of the 4.1 kb ykgK fragment analysed by 2D agarose gel electrophoresis. Coordinates for the SalI restriction fragments detected in previous experiments are given in blue. Coordinates for the BspDI restriction fragment detected by 2D agarose gel electrophoresis are given in purple and the ykgK probe is shown as a black rectangle. The location of the palindrome is shown as a green triangle. The 1.5 kb 3x χ arrays are marked by red lines. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. (B) 2D agarose gel of ΔruvAB ΔrecG mutants containing (DSB+), or not (DSB−), the palindrome and grown in the presence of 0.2% arabinose for 60 minutes. Strains used were DL4260 (lacZ::pal) and DL4313 (lacZ+). (C) Quantification (represented as mean ± SEM where n = 3) of intermediates accumulated in the strain containing the palindrome (DSB+), relative to the strain not containing the palindrome (DSB−) and the percentage of 4-way DNA junctions and 3-way DNA junctions accumulated in each strain.
Figure 6.
Detection of branch migration using PFGE.
(A) Map of the chromosome showing the three SalI fragments around the DSB. The coordinates of the restriction sites are shown in blue. The palindrome is shown as a green triangle and the 1.5 kb 3x χ arrays are shown as red lines. The relative position of probes are represented by small black rectangles. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. (B) Gel of branched DNA retained in the wells of PFGs from DNA isolated from ΔruvC (DL4913 and DL4914) mutants. Samples were run as in Figure 3. (C) Quantifications (represented as mean ± SEM where n = 3) of branched DNA retained in the wells of PFGs from DNA isolated from ΔruvAB (DL4243 and DL4257) mutants (gel shown in Figure 3C) and ΔruvC mutants (gel shown in panel B). Statistical analysis was carried out using a paired T-test. * represents p<0.05, ** represents p<0.01.
Figure 7.
Detection of DNA loss in ΔruvAB ΔrecG and ΔruvC ΔrecG mutants.
(A) Map of the chromosome showing the three SalI fragments surrounding the DSB. The coordinates of the restriction sites are shown in blue. The palindrome is shown as a green triangle and the 1.5 kb 3x χ arrays are shown as red lines. The relative position of the codB probe is represented by a small black rectangle. OP and OD indicate origin-proximal and origin-distal sides of the break, respectively. (B) Gels probed with codB probe and cysN probe. All strains were grown in the presence of 0.2% arabinose for 60 minutes. DSB+ strains contain the palindrome while DSB− strains do not. Strains used were; Rec+ (DL4184 and DL4201), ΔruvAB ΔrecG (DL4260 and DL4313), ΔruvC (DL4913 and DL4914), ΔrecG (DL4311 and DL4312), ΔruvC ΔrecG (DL4941 and DL4942). All lanes shown for each probe were derived from the same membrane. (C) Quantification (represented as mean ± SEM where n = 3) of linear and branched DNA relative to Rec+. Statistical analysis was carried out using an unpaired T-test. ** represents p<0.01.
Figure 8.
(A) Stabilising DSBR intermediates by branch migration. In E. coli, following extensive DNA degradation by RecBCD, a resected 3′ end invades a sister chromosome to establish a D-loop in a reaction catalysed by RecA protein. This is stabilised via branch migration catalysed by RuvAB or RecG to form a Holliday junction that can be resolved to generate a replication fork. Only one end is shown here, but a two-ended reaction can occur as shown in Figure 1. In the absence of branch migration (in a ΔruvAB ΔrecG mutant) the products of RecA-mediated strand-invasion (3-way D-loops) are unstable and non-proficient for repair. This results in extrusion of the invading end from the unbroken chromosome to re-generate a broken end. This end is processed by RecBCD and a second round of strand-invasion is initiated. The whole process is repeated. Over time the broken chromosome is degraded. (B) Stabilising DSBR intermediates by second-end capture. In the canonical eukaryotic DSBR pathway for the repair of a two-ended DSB, one of two 3′ ssDNA ends invades an intact DNA duplex, at a region of homology, to generate a 3-way DNA junction (D-loop). DNA synthesis is then primed off the 3′ DNA end and this leads to the extension of the D-loop, which eventually uncovers enough homology to allow second-end capture. This generates a stable dHJ intermediate, which is then resolved to generate the recombinant products of repair. Alternatively, the 3′ invading DNA strand is extended allowing second-end capture and then both invading strands are ejected and re-anneal in a reaction know as Synthesis Dependant Strand Annealing (SDSA). (C) Unstable DSBR intermediates for the repair of a one-ended DSB by BIR (by D-loop migration) in eukaryotic cells. The 3′ ssDNA ends invades an intact DNA duplex, at a region of homology, to generate a 3-way DNA junction (D-loop). DNA synthesis is primed off the 3′ end. As synthesis proceeds, the unstable D-loop migrates with the replication fork, resulting in the extrusion of the newly synthesised strand and conservative DNA replication. Template switching may occur. The reaction ends when the D-loop either reaches the end of a chromosomes or converges with an oncoming replication fork.