Short-range translocation by a restriction enzyme motor triggers diffusion along DNA

Cleavage of bacteriophage DNA by the Type III restriction-modification enzymes requires long-range interaction between DNA sites. This is facilitated by one-dimensional diffusion (‘DNA sliding’) initiated by ATP hydrolysis catalyzed by a superfamily 2 helicase-like ATPase. Here we combined ultrafast twist measurements based on plasmonic DNA origami nano-rotors with stopped-flow fluorescence and gel-based assays to examine the role(s) of ATP hydrolysis. Our data show that the helicase-like domain has multiple roles. First, this domain stabilizes initial DNA interactions alongside the methyltransferase subunits. Second, it causes environmental changes in the flipped adenine base following hydrolysis of the first ATP. Finally, it remodels nucleoprotein interactions via constrained translocation of a ∼ 5 to 22-bp double stranded DNA loop. Initiation of DNA sliding requires 8–15 bp of DNA downstream of the motor, corresponding to the site of nuclease domain binding. Our data unify previous contradictory communication models for Type III enzymes.

Release of prebound enzyme from its target site on oligoduplexes (named according to Supplementary Fig. 13) following mixing with heparin and with (black) or without (magenta) ATP, measured using stopped flow fluorescence anisotropy.Data from two independent repeats are shown.
Binding and loop translocation events are not observed in the absence of an EcoP15I recognition site.a-b, Representative time trajectories with ATP (w/ ATP) of the angular position of the nano-rotor (grey, at 4000 Hz; red, after 100-point sliding average ≜ Hz) without (w/o) EcoP15I (panel a) or with (/w) 4.66 nM EcoP15I (panel b).A state approximation of the trajectories was determined using vbFRET (vbFRET_June10) 61 (blue).
Nano-rotor angle (rad) unwound DNA (bp) Nano-rotor angle (rad) unwound DNA (bp) Supplementary Fig. 3: Analysis of the lifetimes of free and bound states of the EcoP15I-DNA interaction.Representative time trajectories of the angular position of the nano-rotor (grey, at 4000 Hz; red, after 100-point sliding average ≜40 Hz).A two-state approximation of the trajectories was determined using vbFRET (vbFRET_June10) 61 (blue).The analysis was done for: a, control measurement without EcoP15I (w/o EcoP15I); b, measurements with EcoP15I but without ATP (w/ EcoP15I, w/o ATP); and c, measurements with both EcoP15I and ATP (w/ EcoP15I, w/ ATP).Indicated are examples for the times spent in the free state (tfree) and the "bound" state (tbound) used to determine lifetime distributions displayed in Fig. 1f.Characterisation of cyanine dye-labelled EcoP15I and DNA.a, The DNA-Cy5 substrate (Fig. 2b) was made by annealing 2AP_38_Fwd and 38_Rev (Supplementary Table 4).Steadystate fluorescence excitation (magenta) and emission (black) spectra for the Cy5 label in 25 nM DNA-Cy5 (left) and the Cy3 label in 75 nM EcoP15I 339-Cy3 (right).For excitation spectra, the wavelength was set to 570 nm for Cy3 and 670 nm for Cy5.For emission spectra, the excitation wavelength was set to 546 nm for Cy3 and 600 nm for Cy5.b, Steady-state fluorescence Cy5 emission spectra for 25 nM DNA-Cy5 and where indicated 75 nM unlabelled EcoP15I and 4 mM ATP. Excitation wavelength was set to 600 nm.c, Steady-state fluorescence Cy3 emission spectra for 75 nM EcoP15I 339-Cy3 and, where indicated, 25 nM unlabelled 38/38 oligoduplex and 4 mM ATP. Excitation wavelength was set to 546 nm.d, Steady-state fluorescence emission spectra for 75 nM EcoP15I 339-Cy3 and DNA-Cy5 mixed with 4 mM ATP, as indicated.Excitation wavelength was set to 546 nm except for the free DNA trace which was set to 600 nm.e, Cleavage activity assays for 50 nM Wild Type EcoP15I (left) or EcoP15I 339-Cy3 (right) and 2 nM pKA16.Time points were separated on a 1% (w/v) agarose gel (representative gel of 2 repeats).Time points were 0, 0.5, 1, 2, 4, 8, 16, 32 and 64 min.As well as the supercoiled (SC) substrate, nicked open circle (OC) intermediate and linear (LIN) product, a band corresponding to a DNA-protein band-shift was also observed.loop translocation properties of cyanine dye-labelled EcoP15I are similar to wild type EcoP15I.a, Representative time trajectory of the angular position of the nano-rotor (grey, at 4000 Hz; red, after 100-point sliding average ≜ 40 Hz) with 4.66 nM EcoP15I 339-Cy3 without ATP (w/o ATP).Reversible positive rotational shifts of 0.7 ± 0.07 rad (1 ± 0.1 bp) were observed (the EcoP15I 339-Cy3 "DNA-bound" state).b, With ATP (w/ ATP), sawtooth-like loop translocation events were detected (blue asterisk).c, Representative examples of different loop translocation events including color coded identification for different EcoP15I 339-Cy3-DNA interaction states (grey, free state; red, bound state; blue, loop translocation).d, Maximum loop size, loop translocation rate and loop translocation time (∆tloop translocation), with mean values of: 3.8 ± 0.1 bp, 9.2 ± 1.5 bp/s and 320 ± 23 ms, respectively (N = 42, error S.E.).e-g, Plots of the loop translocation rate vs. loop size (panel e), ∆tloop translocation vs. loop size (panel f) and the loop translocation rate vs. ∆tloop translocation (panel g) plotted for individual events.kbind,Cy3 = -3.02± 0.02 /s k bind,Cy5 = 3.13 ± 0.01 /s 7 Supplementary Fig. 6: Fluorescence changes due to DNA binding and ATP-dependent dissociation of cyanine dye-labelled EcoP15I from DNA. a, Stopped flow fluorescence measurement of binding of 75 nM EcoP15I 339-Cy3 and 25 nM DNA-Cy5.Excitation wavelength was set to 546 nm and the emission of Cy3 (red, right y-axis) and Cy5 (blue, left y-axis) measured simultaneously.The dashed and dotted lines are single exponential fits to give the rate constants shown (errors S.E.M.).b, Stopped flow fluorescence measurement of the ATP-dependent dissociation of 75 nM unlabelled EcoP15I from 25 nM DNA-Cy5, with final concentrations of 4 mM ATP and 2.5 µM heparin trap.Excitation wavelength was set to 600 nm and the emission of Cy5 (blue) was measured.c, Stopped flow fluorescence measurement of the ATP-dependent dissociation of 75 nM EcoP15I 339-Cy3 from 25 nM unlabelled 38/38 oligoduplex (see Supplementary Fig. 13) , with final concentrations of 4 mM ATP and 2.5 µM heparin trap.Excitation wavelength was set to 546 nm and the emission of Cy3 (red) measured.
λ em,max = 377 nm DNA + EcoP15, λ em,max = 369 nm DNA + EcoP15I + ATP, λ em,max = 370 nm DNA + EcoP15I + ATPγS, λ em,max = 371 nm 21.9 ± 1.1 /s (71.3 ± 2.4%) k slow = 4.2 ± 0.4 /s 8 Supplementary Fig.7: Fluorescence changes associated with EcoP15I association with a 2aminopurine labelled oligoduplex.a, Steady-state 2-aminopurine fluorescence emission spectra for 500 nM 2-Ap oligoduplex (Fig.2d) made by annealing 20/50_P15F_11Cy5 and 20/50_P15R (Supplementary ATP-dependent DNA dissociation of EcoP15I as a function of downstream DNA length measured using stopped flow anisotropy. loop translocation properties of wild type EcoP15I at 6 pN stretching force are similar to those obtained at 3 pN stretching force.a, Representative examples of different loop translocation events including color coded identification for different wild type EcoP15I-DNA interaction states (grey, free state; red, bound state; blue, loop translocation) at an applied external force of ~6 pN.b, Maximum loop size and loop translocation rate exhibiting bimodal distributions, with mean values of: 4.3 ± 0.1 bp, 23.4 ± 1.2 bp and 10.0 ± 0.6 bp/s, 23.0 ± 0.8 bp/s, respectively (error S.E.).The loop translocation time (∆tloop translocation) exhibits a monomodal distribution with a mean value of 264 ± 10 ms (N = 77, error S.E.).c, Plots of the loop translocation rate vs. loop size (left), ∆tloop translocation vs. loop size (middle) and the loop translocation rate vs. ∆tloop translocation (right) plotted for individual events.The effect of DNA stretching force on the initiation of sliding by EcoP15I.a, Schematic of the C-Trap flow cell.DNA tethers were formed in Channels 1-3 before binding labelled enzyme (655 nm quantum dot-streptavidin-biotinEcoP15I) to a single central recognition site in Channel 4. To initiate sliding, the tether was moved from a fixed waypoint in Channel 3 containing only imaging buffer to a waypoint in Channel 5 containing buffer plus 4 mM ATP.During this transit period the precise ATP concentration is undefined, and the tether is displaced from the kymograph axis by hydrodynamic drag such that enzyme position cannot be determined, resulting in a dead time of 1.3 s. b, Representative kymographs at either 1.5 or 15 pN showing an initial period where EcoP15I remains bound at the site, after which sliding initiates to produce 1-D diffusive motion.Zero time is defined as the start of the transit period, with the dotted line showing when the waypoint in Channel 5 is reached.Occasional gaps in the traces are due to quantum dot blinking.c, Sliding initiation times determined from kymographs (N = 16 for 1.5 pN and N = 20 for 15 pN) represented as inverted cumulative probability plots fitted to exponential functions to give the initiation time (τ) (Standard Error).Uncertainty due to the transit dead time is smaller than the diameter of data points.d, Representative kymographs showing the positions of site-bound EcoP15I with or without 4 mM ATPS while changing the DNA stretching force (1.5, 5, 10, 15, and 20 pN) (N = 19 and 12 for Buffer Only and ATPγS, respectively).Bead autofluorescence is visible at the upper and lower kymograph edges.Occasional gaps in the traces are due to quantum dot blinking.In each case, EcoP15I remained bound at the central site throughout, and elevated forces (up to 50 pN) did not induce 1-D diffusion.
EcoP15I binding to oligoduplexes.a, Band-shift assays using 0.1 nM 32 Plabelled oligoduplex with 17 bp downstream DNA, without (w/o) or with (w/) pre-bound streptavidin mixed with varying concentrations of EcoP15I as indicated.The free and bound forms were separated by native polyacrylamide electrophoresis (representative gel of 3 repeats).b, The mean of the quantified data (N = 3, error bars S.D.) were fit using the tight binding equation (Kd, error S.E.).c, Band-shift assays using 0.1 nM 32 P-labelled oligoduplexes with either 10 bp (10/38) or 2 bp (2/38) upstream DNA mixed with varying concentrations of EcoP15I as indicated.The free and bound forms were separated by native polyacrylamide electrophoresis (representative gel of 3 repeats).d, The mean of the quantified data (N = 3, error bars S.D.) were fit using the tight binding equation (Kd, error S.E.).Examples of DNA substrate nomenclature.The DNA recognition site is highlighted in orange and the region contacted by the helicase domain is highlighted in pink.Each substrate contains six fixed residues of DNA upstream of the DNA recognition site and a variable number of base pairs downstream.The naming system is "X/Y", where "X" and "Y" represent the lengths in nucleotides downstream of the target site of the MS and TS, respectively.a, Sequence of the 50 bp "full length" DNA substrate 38/38 on which EcoP15I is known to have normal activity and examples of substrates with truncated downstream dsDNA.b, Examples of substrates with 5ʹ or 3ʹ overhangs downstream of the DNA recognition site used to investigate the effect of strand polarity on EcoP15I activity.

Table 4
) and, where indicated, 600 nM EcoP15I, 4 mM ATP and 4 mM ATPγS.Excitation wavelength was set to 311 nm.b, Stopped flow fluorescence measurement at 311 ± 3 nm and using a 360 nm long-pass filter of 125 nM 2-Ap oligoduplex mixed with either 150 nM EcoP15I or a buffer control.The association curve was fitted with a double exponential (solid black line) to give the parameters shown (errors S.E.M.).

Table 3 : Sequence of dsDNA fragment used as a PCR template to make the substrate in Fig. 3 and Extended Data Fig. 4.
Sites for EcoP15I (CAGCAG), EcoPI (AGACC) and LlaGI (CTNGAYG) are underlined and in bold font.DNA fragment synthesised by Integrated DNA Technologies.

Table 5 : Oligodeoxyribonucleotide sequences annealed to make substrates for stopped
-flow spectroscopy.Oligodeoxyribonucleotides synthesised by Integrated DNA Technologies.