Single-molecule live-cell imaging visualizes parallel pathways of prokaryotic nucleotide excision repair

Abstract

In the model organism Escherichia coli, helix distorting lesions are recognized by the UvrAB damage surveillance complex in the global genomic nucleotide excision repair pathway (GGR). Alternately, during transcription-coupled repair (TCR), UvrA is recruited to Mfd at sites of RNA polymerases stalled by lesions. Ultimately, damage recognition is mediated by UvrA, followed by verification by UvrB. Here we characterize the differences in the kinetics of interactions of UvrA with Mfd and UvrB by following functional, fluorescently tagged UvrA molecules in live TCR-deficient or wild-type cells. The lifetimes of UvrA in Mfd-dependent or Mfd-independent interactions in the absence of exogenous DNA damage are comparable in live cells, and are governed by UvrB. Upon UV irradiation, the lifetimes of UvrA strongly depended on, and matched those of Mfd. Overall, we illustrate a non-perturbative, imaging-based approach to quantify the kinetic signatures of damage recognition enzymes participating in multiple pathways in cells.

Fig. 1. Nucleotide excision repair in Escherichia coli.

Damage detection in nucleotide excision repair in E. coli proceeds via global damage surveillance executed by UvrA₂(B), and RNA polymerase (RNAP) transcribing damaged template DNA. The UvrA dimer loads UvrB, which verifies the presence of DNA damage in a strand-specific manner. Alternately, stalled elongation complexes at the site of DNA damage are rescued by the transcription-repair coupling factor Mfd, which in turn recruits UvrA₂(B) to the site of the stalled RNAP. This is followed by strand-specific loading of UvrB at the site of the lesion. Following damage verification by UvrB, a single-stranded patch of DNA containing the damage is incised by the UvrC endonuclease. This is followed by repair synthesis and ligation coordinated by UvrD, PolI and LigA.

Fig. 2. Construction of uvrA-YPet and imaging of UvrA-YPet.

a A chromosomal fusion of UvrA to the yellow fluorescent protein (YPet) under the native uvrA locus was created using λ Red recombination in MG1655 cells. In a second approach, the uvrA-YPet allele was expressed under the native uvrA promoter from a low-copy plasmid in ΔuvrA cells. b Cells expressing fluorescent UvrA-YPet were grown to early exponential phase and loaded in a flow cell. Cells were imaged under constant supply of aerated growth medium. c Fluorescence images of UvrA-YPet reveal a mixture of foci and diffuse cellular background signal. Scale bar represents 5 μm. Cell outlines are provided as a guide to the eye. d Schematic of interval imaging approach employed to measure the off rates of fluorescently tagged proteins in cells. Each acquisition is collected in two phases. In the first phase, fluorescent signal is bleached to enable observation of single fluorescent YPet molecules. In the second phase, a dark frame τ_d is introduced such that the time-lapse time τ_tl = τ_d + τ_int, where τ_int is the integration time (100 ms). In this phase, the lifetimes of individual binding events of UvrA-YPet molecules are measured and combined to obtain a cumulative residence time distribution.

Fig. 3. DNA-binding lifetimes of UvrA-YPet in the presence of UvrB.

a Kinetics of UvrA-YPet interactions with DNA can be detected in the absence of UvrB and Mfd, in ΔuvrA ΔuvrB Δmfd cells expressing UvrA-YPet from a low-copy plasmid or in ΔuvrB Δmfd cells expressing UvrA-YPet from the chromosome. b Cartoon illustrates DNA binding by the mutant UvrA(Δ131–250)-YPet, which is defective in interacting with UvrB and Mfd. c Bar plots represent lifetimes of DNA-bound UvrA-YPet (plasmid: n = 34,927 counts from five repeats; chromosomal: n = 9703 counts from three repeats) and mutant UvrA(Δ131–250)-YPet (a total of n = 88,232 counts from four repeats) in the corresponding genetic background. Lifetimes were obtained from globally fitting the cumulative residence time distributions (CRTDs) (see also Supplementary Fig. 2a–f). Where two kinetic sub-populations are detected, the fast lifetime is displayed in the lower panel. Percentage represents the amplitude of kinetic sub-populations. d Cartoon illustrates loading of UvrB by UvrA. This may occur at sites of undamaged or damaged DNA. e Cartoon illustrates the complex formed by UvrA and the mutant UvrB(ΔβHG) that is deficient in loading reaction. f Bar plots represent lifetimes of DNA-bound UvrA-YPet in Δmfd cells expressing either wild-type UvrB (n = 29,743 counts from 11 repeats) or mutant UvrB(ΔβHG) (n = 16,353 counts from two repeats). Lifetimes were obtained from globally fitting the CRTDs, with more than 1000 counts each CRTD (see Supplementary Fig. 2g–j). Where two kinetic sub-populations are detected, the fast lifetime is displayed in the lower panel. Percentage represents the amplitude of kinetic sub-populations. g Cartoon of the arrested Mfd-UvrA complex. See also Supplementary Fig. 3a, b. Error bars are standard deviations from ten bootstrapped CRTDs. Source data are provided as a Source Data file.

Fig. 4. Dissociation kinetics of UvrA-YPet in TCR-competent cells.

a Lifetimes of UvrA-YPet in uvrA-YPet cells or ΔuvrA/pUvrA-YPet cells untreated or treated with rifampicin. Lifetimes were obtained from globally fitting the CRTDs (see Supplementary Fig. 4a, c, e, g). Long lifetime is presented in black. Short lifetime is presented in red. Percentages represent the amplitude of the slowly dissociating population. b In the presence of UvrB and Mfd, UvrA-YPet in uvrA-YPet cells exhibited a long lifetime of 12.0 ± 0.8 s, reflecting UvrA interactions with UvrB and Mfd (n = 20,111 counts from eight repeats). c Rifampicin treatment abolishes Mfd-RNAP interactions, hence, UvrA-YPet is channelled towards interactions with UvrB, with the long lifetime found to be 9.6 ± 0.6 s (n = 15,355 counts from three repeats). d At eight-fold higher UvrA-YPet concentration obtained upon expression from the low-copy plasmid, the long lifetime of UvrA-YPet in ΔuvrA/pUvrA-YPet cells was found to be 19 ± 1 s, longer than that of UvrA-YPet in uvrA-YPet cells (12 s) (n = 19,853 counts from three repeats). e Upon rifampicin treatment, the long lifetime of UvrA-YPet in ΔuvrA/pUvrA-YPet cells reduced to 11.5 ± 0.6 s (n = 31,788 counts from four repeats). Error bars are standard deviations from ten bootstrapped CRTDs. Source data are provided as a Source Data file.

Fig. 5. Lifetimes of DNA-bound UvrA and Mfd in response to UV irradiation.

a Setup for in situ UV irradiation of cells followed by interval imaging at 30 °C for several hours after UV exposure. b Fluorescence intensity of single uvrA-YPet cells increases following exposure to UV light during the SOS response. c Lifetimes of UvrA-YPet in TCR-deficient cells as a function of time following UV exposure (see Supplementary Fig. 5a–d). The lifetimes at t = 0 min are reproduced from Fig. 4a. (After UV irradiation uvrA-YPet Δmfd all data: n = 21,824 counts from four repeats; 0–25 min: n = 3243; 25–50 min: n = 5175; 50–75 min: n = 6079; 75–100 min: n = 5999 counts). d Lifetimes of UvrA-YPet in uvrA-YPet cells (long: blue, short: red) (after UV irradiation uvrA-YPet all data: n = 25,935 counts from four repeats; 0–25 min: n = 5359; 25–50 min: n = 8070; 50–75 min: n = 7481; 75–100 min: n = 4983 counts) or Mfd-YPet in mfd-YPet cells (green) (after UV irradiation mfd-YPet all data: n = 14,553 counts from four repeats; 0–25 min: n = 5133; 25–50 min: n = 3846; 50–75 min: n = 3119; 75–100 min: n = 2286 counts) as a function of time following UV exposure (Supplementary Fig. 6a–d). The lifetimes at t = 0 min are reproduced from Fig. 4a. Lifetime of Mfd-YPet at t = 0 min is reproduced from our previous work. For b and d, lifetimes were obtained from globally fitting the CRTDs. Lifetimes of the fast and slowly dissociating sub-populations are shown in the lower and upper panels respectively. Dashed lines and error bands represent lifetimes and the corresponding standard deviations obtained from aggregated CRTDs within 100 min following UV exposure. Fitting results are available in Table 2. e Amplitudes of slowly dissociating species of UvrA-YPet in mfd⁺ (blue) or Δmfd (black) cells carrying uvrA-YPet. See also Table 2 for fitting results. f UvrB (orange) controls the release of UvrA (purple) from UvrA-Mfd (green) intermediates. Error bars are standard deviations from ten bootstrapped CRTDs. Source data are provided as a Source Data file.