Dissection of DNA double-strand-break repair using novel single-molecule forceps

Abstract

Repairing DNA double-strand breaks (DSBs) by nonhomologous end joining (NHEJ) requires multiple proteins to recognize and bind DNA ends, process them for compatibility, and ligate them together. We constructed novel DNA substrates for single-molecule nanomanipulation, allowing us to mechanically detect, probe, and rupture in real-time DSB synapsis by specific human NHEJ components. DNA-PKcs and Ku allow DNA end synapsis on the 100 ms timescale, and the addition of PAXX extends this lifetime to ~2 s. Further addition of XRCC4, XLF and ligase IV results in minute-scale synapsis and leads to robust repair of both strands of the nanomanipulated DNA. The energetic contribution of the different components to synaptic stability is typically on the scale of a few kilocalories per mole. Our results define assembly rules for NHEJ machinery and unveil the importance of weak interactions, rapidly ruptured even at sub-picoNewton forces, in regulating this multicomponent chemomechanical system for genome integrity.

Figure 1. Double-strand break repair by NHEJ proteins at single-molecule resolution

(A) A 600 bp dsDNA segment (magenta) joins two 1.5 kbp dsDNA segments (blue, red), forming a construct in which two blunt ends face each other. Stars represent phosphate groups. The construct is tethered to a treated glass surface and a 1-micron magnetic bead. Magnets located above the sample generate a controlled extending force on the DNA (green arrow), and the DNA end-to-end extension is determined in real-time. Ligation is observed via a series of four steps: (1) at high force a high-extension state is initially observed, (2) the force is lowered allowing the DNA ends to interact, (3) the force is returned to its initial value but if end ligation has occurred the construct cannot recover its initial extension, (4) the initial extension is recovered upon specific cleavage of repaired DNA. (B) Time-trace obtained upon force-modulation (red) in the presence of Ku, DNA-PKcs, PAXX, XLF, XRCC4 and Ligase IV. Initially, DNA extension (blue points) is shown to alternate between a low and a high value upon force modulation. After addition of NHEJ components (black up arrow) the maximum extension displayed by the construct is reduced. After washing the sample with 0.2% SDS (break in time-trace), addition of SmaI (red up arrow) results in an increase, ∆l in DNA extension. (C) Histogram of ∆l values. Red line is a fit to a Gaussian distribution, with a maximum at 161 ± 8 nm (SD, n=28 cleavage events). (D) DNA ligation probability per traction cycle. We monitored 50 DNA molecules; of 36 molecules repaired 28 were monitored throughout the cleavage reaction. Red line: single-exponential fitting yields a time constant of ~0.8 ± 0.2 cycles (SEM, n=36) or ~175s.

Figure 2. Ku, DNA-PKcs and PAXX are necessary to stabilize DNA-end synapsis

(A) Experimental design. DNA is prepared with blunt ends using SmaI digest, stars represent phosphate groups. (B) Representative time-trace obtained upon application of the force-modulation pattern (red). DNA is prepared with blunt ends using SmaI digest. Inset shows an expanded view of an end-interaction rupture event, which can be characterized by both the change in DNA extension upon rupture, Δl, and the duration of the synaptic event prior to rupture, t_synapsis. (C) Histogram of DNA extension change, ∆l, upon rupture event. Red line is a fit to a Gaussian distribution, with a maximum (red arrow) at 166 ± 10 nm (SD). The entire histogram contains n=129 events, of which 98 are within three standard deviations from the peak. (D) Lifetime distribution of the synaptic state is fit to a single-exponential distribution (red line), giving a lifetime of 2.2 ± 0.3 s (SEM, n=98).

Figure 3. DNA synapsis on the 2-second timescale in the presence of Ku, DNA-PKcs, and PAXX: control experiments

(A) Synapsis by Ku, DNA-PKcs and PAXX does not require phosphorylated DNA ends. (Left) Time-trace showing rupture events. (Middle) Amplitude distribution of rupture events (n=69). Distribution peak is fit to a Gaussian (red line) with a maximum at 163 ± 2 nm (SEM; σ=12 nm; n=40). The distribution displays additional density for Δl values of 75 nm and 400 nm. Values of 400 nm are consistent with the extension of a 1500 bp DNA segment at the force employed, and thus likely correspond to “loop-back” interactions between the tip of a 1500 bp DNA segment and the surface to which that segment is anchored. The smaller peak is consistent with local bending or wrapping deformations of DNA with ~50 nm persistence length [37]. (Right) Lifetime distribution of synaptic events follows a single-exponential distribution (red line) with a mean of 1.9±0.5s (SEM, n=40). (B-E) Representative time-traces show that the combinations of (B) Ku, DNA-PKcs, and the PAXX mutant; (C) Ku, DNA-PKcs, and XLF; (D) Ku and DNA-PKcs; and (E) PAXX alone do not lead to 2-second synapsis. The well-known contamination of DNA-PKcs preparations by Ku [25, 38] means it is not formally possible to test DNA-PKcs alone.

Figure 4. Maximum stabilization of DNA end synapsis by Ku, DNA-PKcs, PAXX, XLF, XRCC4 and Ligase IV

(A) Experimental design. DNA is prepared with blunt ends using SmaI digest and then dephosphorylated (see Materials & Methods). (B) Representative time-trace obtained upon application of the force-modulation pattern (red) in the presence of Ku, DNA-PKcs, PAXX, XLF, XRCC4 and Ligase IV. Transient synapsis is observed as an intermediate DNA extension state, which spontaneously reverts to the maximum extension, allowing us to characterize the duration of the intermediate state and the change in extension upon reversion to the maximum extension. (C) Histogram of DNA extension change (∆l) upon rupture event. Red line is a fit to a Gaussian distribution, with a maximum (red arrow) at 165 ± 9 nm (SD, n=324 events in the histogram, of which 183 within three standard deviations from the peak). (D) Lifetime distribution of the synaptic state is fit to a single-exponential distribution (red line), giving a lifetime of 66.4 ± 7.6 s (SEM, n=183).

Figure 5. Model for multivalent stabilization of DNA-end synapsis by NHEJ machinery

Schematic model for synapsis and repair involves (top) an initial synaptic complex formed by two DNA-PK holoenzymes but with a lifetime in the range of hundreds of milliseconds and which can be stabilized by incorporation of PAXX (left path) or XRCC4–XLF–Ligase IV (right path). A complete complex stabilized by both PAXX and XRCC4–XLF–Ligase IV has the longest lifetime (bottom) and leads to efficient ligation of the DSB.