Phosphorylated CtIP bridges DNA to promote annealing of broken ends

Significance The DNA of our cells is constantly exposed to various types of damaging agents. One of the most critical types of damage is when both strands of the DNA break, and thus such breaks need to be efficiently repaired. It is known that CtIP promotes nucleases in DNA break repair. Here we show that CtIP can also hold the two DNA strands together in solution when DNA is free to move, using novel methodology that allows the monitoring of thousands of single DNA molecules in nanofabricated devices. DNA bridging likely facilitates the enzymatic repair steps and identifies novel CtIP functions that are crucial for repairing broken DNA.


Single Molecule Nanofluidics
The various CtIP derivatives were mixed with sticky-ended l-phage DNA (48502 bp, Roche) at the different ratios indicated in each experiment in buffer I (10 mM Tris-HCl pH 7.6, 10 mM NaCl and 5 mM DTT). Incubation of 330 nM CtIP per DNA-end was used for most experiments, where the DNA base pair concentration was set to 4 µM. This corresponds to an approximate ratio of one CtIP tetramer to 50 bp DNA. For the structural mutants, the ratio of protein to DNA was multiplied by two for CtIPL27E (dimer) and by four for CtIPD160 (monomer). For the plasmid sample, 4 µM DNA was mixed with 330 nM and 660 nM wtCtIP respectively. For blunt-ended T7-DNA (39936 bp, Mabion), 4 µM DNA was mixed with 400 nM wtCtIP, corresponding to 500 tetramers per DNA end.
4µM of PciI-digested l-DNA was mixed with 330 nM wtCtIP and CtIPL27E. The Sae2-derivatives were mixed with DNA at a ratio of 1 protein to 1, 25 and 50 base pairs respectively. The samples were incubated at 37°C for one hour followed by addition of the bis-intercalating fluorescent dye YOYO-1 at a ratio of 1 dye molecule per 5 bp DNA and incubation for additional 10 minutes at 25°C. The samples were diluted to a final concentration of 1 µM DNA base pair and supplemented with sodium dodecyl sulphate (SDS) at a final concentration of 0.05% (w/w) to obtain buffer A (10 mM Tris-HCl pH 7.6, 10 mM NaCl, 5 mM DTT, 0.05% SDS), which was used as analysis buffer in all nanofluidic experiments. SDS was included to reduce sticking of the DNA-protein complexes to the channel walls.
The silicon dioxide based nanofluidic chips with channel dimensions of 150 x 100 nm 2 were fabricated as described elsewhere 1 and used for all single-molecule experiments. The nanofluidic chip was equilibrated with buffer A at 25°C prior to loading of the YOYO-1 stained DNA-CtIP sample in one of the four reservoirs. Pressurized nitrogen gas was used to control the liquid flow within the nanofluidic chip. By manipulating the applied pressure on each individual reservoir, the pre-formed DNA-CtIP complexes were pre-concentrated in the microchannel at the entrance of the nanochannels, before being driven in to the nanochannels, to simultaneously image as many individual DNA-CtIP complexes as possible. The confined DNA-CtIP complexes were visualized on an inverted fluorescence microscope (Zeiss AxioObserver.Z1) equipped with a 100x oil immersion objective (NA = 1.46), a Colibri 7 LED light source (Zeiss) and an sCMOS Prime 95B camera (Photometrics). Blue light (469/38 nm) was used to excite the sample and the emission was passed through a single band pass filter (530/30 nm) before reaching the detector. To determine the extension of the DNA-CtIP complexes, 50 frames were recorded for each molecule, with a frame rate of 0.13 s/frame. For visualizing DNA unfolding, up to 300 frames were recorded at the same frame rate.

Data analysis
The collected images were analyzed using a custom-written MATLAB-interfaced software, where each individual DNA-CtIP complex was detected. The recorded image stacks for each detected complex were converted to a kymograph, from which the extensions and corresponding standard deviations were calculated 2 . Complexes, which were found to break during the course of recording were excluded from the size analysis. Size distribution histograms and standard deviation scatterplots were created using MATLAB.
To distinguish the circular DNA-CtIP complexes from the other populations of molecules, a clustering approach was employed using the free statistical software R. The molecule extensions and the associated internal standard deviations were log-transformed, followed by applying hierarchical clustering using Euclidian distance metric and employing Ward's minimum variance method to specify the dissimilarity of clusters. The complexes found in the cluster corresponding to the circular fraction are highlighted in the respective scatter plots (molecule extension vs standard deviation). A similar approach was used to distinguish the full-size linear l-DNA-CtIP complex. In this way the more scattered full-size linear DNA-CtIP fraction formed distinct clusters, from which we could determine the number of intact l-DNAs as well as the mean extension and the corresponding standard deviation. In order to account for all full-size linear complexes in the population, which display higher lateral flexibility than the circular complexes, the sizes of the partially clustered linear complexes were plotted in a histogram to which a normal probability distribution function was fitted. The resulting mean extension (µ) and standard deviation (s) were obtained from the fit. All molecules with an extension of µ ± 2s were considered to be full-size linear l-DNA-CtIP complexes, independent of standard deviation.
The circularization efficiency (Ec) was calculated from the number of circles detected, compared to the total number of circular molecules, full-length l-DNA molecules and longer complexes (corresponding to blue, black and red fractions, respectively, in the scatterplots and histograms). The level of concatemerization was calculated from the number of molecules longer than that of one linear full-length l-DNA molecule, compared to the total number of molecules corresponding to circular, linear full-length and larger molecules. DNA fragments were not included in the calculations.
For the sequence specificity analysis, kymographs of linear DNA molecules were picked based on their fluorescence emission signal-intensity profiles. Molecules with profiles resembling the GC-content profile of l-DNA, were combined to form a consensus intensity-profile 3 . The median relative intensity of the GC-and AT-rich regions (first and second half of the intensity profile respectively) were determined and the difference was calculated for each sample.

Atomic force microscopy (AFM)
4 µM bp of a pET-plasmid was incubated with 330 nM wtCtIP or CtIPL27E. This concentration was set to obtain a ratio of one wtCtIP tetramer or two CtIPL27E dimers per 50 bp DNA, equivalent to that of 500 CtIP tetramers per l-DNA-end in the nanofluidic experiments. The DNA and protein were mixed in buffer I for one hour at 37°C before depositing 15 µl of the sample on a mica surface. The DNA-CtIP complexes were allowed to adsorb to the surface for 10 minutes at room temperature, followed by rinsing the mica with ultrapure MilliQ water. Pressurized nitrogen was used to dry the surface. The AFM images were acquired in air using an NTEGRA Prima scanning probe microscope, operating in tapping mode with golden silicon probes (force constant 1.45-15.1, resonance frequency 87-203 kHz). The scanning rate was 1 Hz. The resulting images were treated in the open source software Gwyddion 4 .

l-phosphatase treatment
Phosphorylated wtCIP was dephosphorylated by mixing 1 µg of the protein with 200 U lphosphatase (NEB) in 1x PMP buffer (NEB) and 1mM MnCl2. The total volume was adjusted to 20 µl by addition of water. A mock-reaction was run simultaneously, where the l-phosphatase was replaced by water. The samples were incubated at 30°C for 15 minutes, followed by separation on a denaturing polyacrylamide gel.

Electrophoretic mobility shift assay
When linear DNA was used, pUC19 was linearized with EcoRI (New England Biolabs) according ATPγS. Topoisomerase I was added to relax the circular DNA, and the reaction was incubated for 10 min at 37°C. After the addition of wtCtIP, the reactions were incubated on ice for 30 min. Loading dye (50% glycerol, bromophenol blue) was then added and the products were separated by 0.6% agarose gel electrophoresis in Tris-Acetate-EDTA (TAE) buffer. The electrophoresis was carried out in a cold room at 4°C and the DNA was visualized by staining with GelRed (Biotium).

Nuclease assay
To prepare the quadruple blocked 70-bp long DNA substrate, PC210 and PC211 oligonucleotides were used, as described previously 5 . Briefly, PC210 was labeled at the 5ꞌ-end by T4 polynucleotide kinase (New England Biolabs) and [γ-32P] ATP (Perkin Elmer) according to the manufacturer's instructions. Unincorporated nucleotides were removed using Micro Bio-Spin P-30 Tris chromatography columns (Biorad) 6 .
Endonuclease assays (15 μl volume) were performed in nuclease buffer containing 25 mM Tris-HCl pH 7.5, 5 mM magnesium acetate, 1 mM manganese acetate, 1 mM dithiothreitol (DTT), 1 mM ATP, 0.25 mg/ml BSA (New England Biolabs), 1 mM phosphoenolpyruvate (Sigma), 80 U/ml pyruvate kinase (Sigma), and 1 nM oligonucleotide-based DNA substrate (in molecules) 6 . The reactions were supplemented with 15 nM streptavidin (Sigma) and incubated for 5 min at room temperature to block the biotinylated ends of the DNA substrates. The recombinant proteins were then added to the reactions on ice and samples were incubated at 37°C for 30 min. After the addition of 0.5 μl of 0.5 M EDTA and 1 μl Proteinase K (19 mg/ml, Roche), reactions were stopped by incubation at 50°C for 30 min. Finally, 16.5 μl loading buffer (5% formamide, 20 mM EDTA, bromophenol blue) was added to all samples and the products were separated on 15% polyacrylamide denaturing urea gels (19:1 acrylamide-bisacrylamide, Bio-Rad), as described elsewhere 6 . The gels were fixed in fixing solution (40% methanol, 10% acetic acid, 5% glycerol) for 30 min at room temperature and dried on a 3MM Chr paper (Whatman). The dried gels were exposed to storage phosphor screen (GE Healthcare) and scanned by a Typhoon Phosphor Imager (FLA 9500, GE Healthcare).  (c, d). Only a slight shift is observed for the circular DNA, which may suggest higher affinity of wtCtIP for circular compared to linear DNA.

Counts
Ext ns on Ext ns on