Studying the organization of DNA repair by single-cell and single-molecule imaging

Abstract

DNA repair safeguards the genome against a diversity of DNA damaging agents. Although the mechanisms of many repair proteins have been examined separately in vitro, far less is known about the coordinated function of the whole repair machinery in vivo. Furthermore, single-cell studies indicate that DNA damage responses generate substantial variation in repair activities across cells. This review focuses on fluorescence imaging methods that offer a quantitative description of DNA repair in single cells by measuring protein concentrations, diffusion characteristics, localizations, interactions, and enzymatic rates. Emerging single-molecule and super-resolution microscopy methods now permit direct visualization of individual proteins and DNA repair events in vivo. We expect much can be learned about the organization of DNA repair by linking cell heterogeneity to mechanistic observations at the molecular level.

Keywords: Cell heterogeneity; DNA damage responses; DNA repair; Single-cell imaging; Single-molecule fluorescence; Super-resolution microscopy.

Fig. 1

Sources of heterogeneous protein levels in genetically identical cells. Green circles represent a fluorescent reporter protein. (A) Gene expression noise: Protein levels fluctuate due to stochastic gene expression bursts and synthesis of multiple proteins per mRNA molecule. At time “t”, an example cell (red) has one mRNA copy and a low protein level, while the other cell (blue) has two mRNA copies and a higher protein level. (B) Random partitioning: Cell division occurs with different protein numbers in the two daughter cells. (C) Gene activation/repression: Following an induction signal, the timing of gene activation (or repression) is dictated by stochastic unbinding (or binding) events of a transcription repressor. (D) Bistability: Strong positive feedback regulation produces bistable behavior in which random excursions above an expression threshold trigger a complete switch in protein levels. (E) Pulsed feedback: Combination of positive and negative feedback regulation causes pulsed gene expression at varying amplitude and/or frequency. (F) The SOS response is an example for a pulsed DNA damage response: In the absence of damage, LexA dimers repress transcription of SOS genes. DNA damage triggers cleavage of LexA by active RecA, leading to pulses of SOS induction in a single cell (curve adapted from Ref. [18]).

Fig. 2

Methods for measuring protein mobility in vivo. (A) FRAP: The characteristic recovery time of the FRAP curve after bleaching reports on the protein mobility and exchange rate of molecules at binding sites within the bleaching spot. A difference between the pre- and post-bleach intensities indicates the presence of permanently bound molecules. (B) FCS: The decay time and amplitude of the autocorrelation curve report on the mobility and average concentration of proteins in the focus, respectively. (C) PSF analysis: The motion of a single molecule during the camera exposure time blurs the PSF. Histograms of the PSF width can be used to classify proteins of different mobility. The PSF intensity and sequential photobleaching steps report on the number of fluorescent molecules in a spot. (D) Single-molecule tracking connects localizations of one or few labeled molecules per cell to directly follow their motion. The mean squared displacement (MSD) as a function of the lag time between localizations summarizes the tracking data and distinguishes between immobile, confined, Brownian, or directed motion. (E) Photoactivated single-molecule tracking employs PALM to activate and track arbitrary numbers of labeled molecules per cell in a sequential manner. Protein mobility can be directly classified by the diffusion coefficient per track.

Fig. 3

Measuring single base-excision repair events by DNA polymerase I (Pol1) in live E. coli. Scale bars: 1 μm. Figures adapted from . (A) Photoactivated single-molecule tracking gives a map of Pol1 tracks in a cell with DNA damage by methyl methanesulfonate (MMS). Based on the mean squared displacement (MSD), individual tracks with a low apparent diffusion coefficient are shown in red while tracks of freely diffusing Pol1 are shown in blue. Histograms of the apparent diffusion coefficient can be used to quantify the fraction of bound Pol1 molecules in the absence and presence of MMS damage. (B) The localizations of bound Pol1 molecules (red dots) show that base-excision repair sites are randomly positioned throughout the nucleoid. The histogram shows the distribution of unbound Pol1 molecules across the long cells axis; positions of bound molecules are shown in red. (C) Prolonged treatment with a low dose of MMS for 1 h causes chromosome compaction, as evident from the confinement of tracks to a smaller area compared to cells in panels A and B that were imaged within 20 min of MMS treatment. (D) Individual Pol1 tracks display the search path to find a repair site (light blue), a complete repair event (red), and further diffusion (dark blue). (E) Quantifying the Pol1 damage response during 15 min MMS treatment and subsequent recovery. Using the binding time per repair event and the percentage of bound molecules gives the repair rate per Pol1 molecule. The total repair rate per cell is estimated by counting the number of Pol1 tracks per cell.