Single-molecule imaging reveals multiple pathways for the recruitment of translesion polymerases after DNA damage

Abstract

Unrepaired DNA lesions are a potent block to replication, leading to replication fork collapse, double-strand DNA breaks, and cell death. Error-prone polymerases overcome this blockade by synthesizing past DNA lesions in a process called translesion synthesis (TLS), but how TLS polymerases gain access to the DNA template remains poorly understood. In this study, we use particle-tracking PALM to image live Escherichia coli cells containing a functional fusion of the endogenous copy of Pol IV to the photoactivatable fluorescent protein PAmCherry. We find that Pol IV is strongly enriched near sites of replication only upon DNA damage. Surprisingly, we find that the mechanism of Pol IV recruitment is dependent on the type of DNA lesion, and that interactions with proteins other than the processivity factor β play a role under certain conditions. Collectively, these results suggest that multiple interactions, influenced by lesion identity, recruit Pol IV to sites of DNA damage.

Fig. 1

Particle-tracking PALM imaging of a functional Pol IV-PAmCherry fusion. a Schematic of the fusion of the photoactivatable fluorescent protein gene pamcherry1 to the C terminus of the dinB gene encoding Pol IV (top) and schematic of the relevant Pol IV domains and residues (bottom). b Serial 10-fold dilutions of E. coli strains grown on LB agar plates without (left) and with (right) nitrofurazone (NFZ) added at 8 µM concentration. c Top panels: representative fluorescence micrographs of single activated Pol IV-PAmCherry molecules recorded with 13.3 ms (top left) and 250 ms (top right) integration times, SSB-mYPet foci (bottom left), and a nucleoid labeled with HU-EYFP (bottom right), with overlays of the cell outlines. Bottom panels: the corresponding brightfield micrographs with an overlay of all detected Pol IV-PAmCherry tracks (top left and top right) or all detected SSB-mYPet foci (bottom left) in the cell (Scale bars: 1 µm). d Distributions of the apparent diffusion coefficient (D*) for Pol IV-PAmCherry in undamaged (top) and 100 µM NFZ-treated cells (middle), and the difference (bottom). The dashed lines indicate the threshold D* value for bound molecules (D* < 0.275 µm²/s)

Fig. 2

Cellular localization of Pol IV-PAmCherry and SSB-mYPet in undamaged, NFZ-treated, and MMS-treated cells. Scatter plots of normalized mean coordinates of Pol IV-PAmCherry tracks and the normalized coordinates of SSB-mYPet foci in undamaged (a, b), 100 µM NFZ-treated (c, d), and 100 mM MMS-treated (e, f) cells. Also shown are the projections of the scatter plots along the long (x) and short (y) cellular axes

Fig. 3

Single-cell colocalization of Pol IV-PAmCherry and SSB-mYPet in undamaged, NFZ-treated, and MMS-treated cells. a Distributions of the mean distance between each static Pol IV track and the nearest SSB focus for Pol IV^WT (top) and Pol IV^R,C (top center) in undamaged cells and for Pol IV^WT in cells treated with 100 µM NFZ (bottom center) or 100 mM MMS (bottom). b Radial distribution functions g(r) for the undamaged distance distributions in a. c Radial distribution functions g(r) for the damaged Pol IV^WT distance distributions in a. d Radial distribution functions g(r) for Pol IV^WT in cells treated with 100 µM NFZ for 20 min, 1 h, or 2 h. e Radial distribution functions g(r) for Pol IV^WT in cells treated with 100 mM MMS for 20 min or 1 h. Also shown in panels b–e are random g(r) functions for each data set. Several Pol IV^WT g(r) traces are replotted to enable comparison

Fig. 4

Effect of NFZ and MMS treatment on the number, lifetime, and cellular localization of static Pol IV-PAmCherry tracks. a Distributions of the number of static Pol IV-PAmCherry tracks per cell in undamaged cells and cells treated with 40 µM NFZ, 100 µM NFZ, or 100 mM MMS. The mean of each distribution is indicated by a solid line. b Distribution of the apparent Pol IV-PAmCherry binding lifetime in undamaged cells (blue circles) and cells treated with 100 µM NFZ (red squares) or 100 mM MMS (green stars). Also shown is the apparent Pol IV-PAmCherry binding lifetime in fixed cells (black triangles) as a photobleaching control. c Long-axis cellular localization of Pol IV-PAmCherry in cells treated with 40 µM NFZ

Fig. 5

Long-axis cellular localization profiles and single-cell colocalization analysis of Pol IV mutants. a Long-axis cellular localization of Pol IV^WT and the Pol IV^R,C, Pol IV-D103N, and Pol IV^CD mutants in cells treated with 100 µM NFZ. b Long-axis cellular localization of Pol IV^WT and the Pol IV^R,C and Pol IV-D103N mutants in cells treated with 100 mM MMS. c Radial distribution functions g(r) for Pol IV^WT and the Pol IV^R,C mutant in cells treated with 100 µM NFZ for 20 min. d Radial distribution functions g(r) for Pol IV-SSB distance for Pol IV^WT and the Pol IV^R,C and Pol IV-D103N mutants in cells treated with 100 mM MMS. Also shown in panels c and d are random g(r) functions for each data set. The Pol IV^WT data in a–d are replotted from Fig. 2c, e and Fig. 3c, d to enable comparison

Fig. 6

Effect of NFZ and MMS treatment on RecA cellular localization. Long-axis cellular localization of RecA-GFP foci in undamaged cells (top) and cells treated with 100 µM NFZ (middle) or 100 mM MMS (bottom)