Chi hotspots trigger a conformational change in the helicase-like domain of AddAB to activate homologous recombination

In bacteria, the repair of double-stranded DNA breaks is modulated by Chi sequences. These are recognised by helicase-nuclease complexes that process DNA ends for homologous recombination. Chi activates recombination by changing the biochemical properties of the helicase-nuclease, transforming it from a destructive exonuclease into a recombination-promoting repair enzyme. This transition is thought to be controlled by the Chi-dependent opening of a molecular latch, which enables part of the DNA substrate to evade degradation beyond Chi. Here, we show that disruption of the latch improves Chi recognition efficiency and stabilizes the interaction of AddAB with Chi, even in mutants that are impaired for Chi binding. Chi recognition elicits a structural change in AddAB that maps to a region of AddB which resembles a helicase domain, and which harbours both the Chi recognition locus and the latch. Mutation of the latch potentiates the change and moderately reduces the duration of a translocation pause at Chi. However, this mutant displays properties of Chi-modified AddAB even in the complete absence of bona fide hotspot sequences. The results are used to develop a model for AddAB regulation in which allosteric communication between Chi binding and latch opening ensures quality control during recombination hotspot recognition.


Mass spectrometry
Analysis of the 110 kDa band of interest was performed by the University of Bristol Proteomics service.
Gel bands were subjected to in-gel tryptic digestion using a DigestPro automated digestion unit (Intavis Ltd.). The resulting peptides were diluted 1000-fold and fractionated using an Ultimate 3000 nanoHPLC system in line with an LTQ-Orbitrap Velos mass spectrometer (Thermo Scientific). In brief, peptides in 1% (vol/vol) formic acid were injected onto an Acclaim PepMap C18 nano-trap column (Thermo Scientific). After washing with 0.5% (vol/vol) acetonitrile 0.1% (vol/vol) formic acid peptides were resolved on a 250 mm × 75 μm Acclaim PepMap C18 reverse phase analytical column (Thermo Scientific) over a 150 min organic gradient, using 7 gradient segments (1-6% solvent B over 1min., 6-15% B over 58min., 15-32%B over 58min., 32-40%B over 3min., 40-90%B over 1min., held at 90%B for 6min and then reduced to 1%B over 1min.) with a flow rate of 300 nl min −1 . Solvent A was 0.1% formic acid and Solvent B was aqueous 80% acetonitrile in 0.1% formic acid. Peptides were ionized by nanoelectrospray ionization at 2.1 kV using a stainless steel emitter with an internal diameter of 30 μm (Thermo Scientific) and a capillary temperature of 250°C. Tandem mass spectra were acquired using an LTQ-Orbitrap Velos mass spectrometer controlled by Xcalibur 2.1 software (Thermo Scientific) and operated in data-dependent acquisition mode. The Orbitrap was set to analyze the survey scans at 60,000 resolution (at m/z 400) in the mass range m/z 300 to 2000 and the top twenty multiply charged ions in each duty cycle selected for MS/MS in the LTQ linear ion trap. Charge state filtering, where unassigned precursor ions were not selected for fragmentation, and dynamic exclusion (repeat count, 1; repeat duration, 30s; exclusion list size, 500) were used. Fragmentation conditions in the LTQ were as follows: normalized collision energy, 40%; activation q, 0.25; activation time 10ms; and minimum ion selection intensity, 500 counts.
The raw data files were processed and quantified using Proteome Discoverer software v1.2 (Thermo Scientific) and searched against a combined database consisting of the UniProt Bacillus subtilis 168 database (4217 sequences) plus the AddA and AddB sequences using the SEQUEST algorithm. Peptide precursor mass tolerance was set at 10ppm, and MS/MS tolerance was set at 0.8Da. Search criteria included carbamidomethylation of cysteine (+57.0214) as a fixed modification and oxidation of methionine (+15.9949) as a variable modification. Searches were performed with full tryptic digestion and a maximum of 1 missed cleavage was allowed. The reverse database search option was enabled and all peptide data was filtered to satisfy false discovery rate (FDR) of 5%.

Dye displacement helicase assay
Real-time helicase measurements were performed and analysed as described previously (1). Experiments were performed at 37 o C in a stopped-flow instrument with a xenon-mercury light source (TgK Scientific).
Hoechst 33258 dye was excited at 366 nm through 1 mm slits, and the fluorescence above 400 nm was subsequently recorded. DNA molecules (0.1 nM) were incubated with 1 nM streptavidin in a buffer containing Hoechst 33258 (200 nM), SSB 4 protein (200 nM), BSA (100 mg/ml), tris acetate (25 mM, pH 7.5), magnesium acetate (2 mM) and DTT (1 mM). AddAB enzymes (1 nM), were added to the preformed DNA:streptavidin complexes and incubated at 37 o C for 2 minutes prior to mixing against an equal volume of the same solution but with the streptavidin and DNA substrate omitted and AddA K36A B enzymes (30 nM) and ATP (0.5 mM) added. All concentrations stated are post-mixing. The use of the AddA K36A B mutant prevents re-binding of the enzyme being studied ensuring single-turnover conditions with respect to the DNA substrate (1).

Limited proteolysis with crystallographic substrates
Limited proteolysis experiments were conducted essentially as described in the main text but using crystallographic hairpin DNA substrates with unpaired single-stranded tails. Two substrates were used; Hairpin-Chi+ included a Chi sequence in the correct position for immediate recognition whereas in Hairpin-Chi0 the hotspot sequence was replaced by five thymidine bases (see table below). A single base was included 3' of Chi to mimic the cleaved product post Chi-recognition. In advance of the reaction, substrates were annealed and purified as described previously (2). Mutant AddAB enzymes were expressed and purified as described previously (3). The latch mutants were made in combination with a double nuclease inactivation mutant (D1172A in AddA and D961A in AddB) to mimic the crystallographic conditions that prevent digestion of the exposed single-strand tails. AddAB was preincubated with excess DNA in standard reaction buffer at 37 o C for 10 minutes. The processing of DNA substrates was initiated by the addition of ATP to a concentration of 1 mM. After 20 seconds, α-Chymotrypsin was added to a concentration of 0.5 μg/ml. 10 μL aliquots were removed at indicated time points and added to 10 μL of stop buffer and immediately placed at 95 o C for 2 minutes prior to electrophoresis. Break resection reactions were conducted to confirm that the substrates used in proteolysis experiments behaved similarly to more conventional long substrates with respect to nuclease regulation by Chi. The experiments were performed in standard buffer conditions (see main text), but in the absence of SSB protein, and were run on 8% denaturing polyacrylamide gels to resolve the short resection products.

Oligonucleotides used in this study
The wild type enzyme processes the Chi-containing 73 mer to produce a series of bands that run below the full length substrate DNA (SFigure 3). If the enzyme cuts predominantly 1 ntd upstream of Chi as expected (4), then the predicted Chi-dependent cleavage products should be 71, 63, 54, 46 and 37 nucleotides long. When looking at the lane profile of this reaction distinct peaks occur at these predicted cleavage positions. Furthermore, the Chi recognition mutant AddAB F210A produces the same banding pattern, but these bands are all less intense than with the wild-type (6% vs 17% of total input DNA). This indicates that these bands reflect cleavages made in the DNA in response to Chi. When Chi is absent in the "scrambled" control substrate, there are no ATP-dependent bands that run below the substrate.
AddAB has presumably unwound the DNA without degrading it because the lifetime of the unwinding event is much smaller than the time required to cleave the DNA. This is as expected based on the infrequent cleavage pattern observed on much longer substrates (5).

Supplementary discussion 2 -The proteolysis product of AddAB* maps to the helicase-like domain of AddB and close to the latch binding pocket.
Evidence for a conformational change upon Chi recognition was provided by a change in the products of limited proteolysis. To gain further structural insight into the nature of this change, we initially attempted to map the Chi-dependent proteolytic site using N-terminal sequencing, but this repeatedly failed due to a lack of material. We subsequently turned to Orbitrap mass spectrometry to identify the peptide products Under the identical conditions for the experimental samples, we obtained very low coverage of AddA, but consistently obtained peptides that covered much of the C-terminal region of AddB. The peptides mapped from as early as residue 265 to near the C-terminus. Removal of 265 residues from the N-terminus produces a protein of theoretical mass 103 kDa, which is within error of our estimate for the molecular mass of the proteolysis product (~110 kDa) from molecular weight markers. The possible cleavage region is within the helicase-like domain of AddB and includes the latch binding pocket. The results are therefore consistent with, but do not prove, movement of the ionic latch structure upon Chi recognition.
Supplementary discussion 3 -The latch mutant displays similar biochemical properties to Chimodified wild type AddAB.
The modification of AddAB by Chi can be monitored by at least three biochemical hallmarks: attenuation of 3-directed nuclease activity (4), reduction in translocation rate (6,7) and stimulation of helicase activity at limiting [SSB] (1). Experiments in the main paper using triplex displacement assays show that the latch mutant has translocation properties that partially resemble the Chi-modified wild type even in the absence of Chi sequences. To corroborate the triplex displacement results we measured helicase activity in real-time as described in (1)