Assembly dynamics of supramolecular protein-DNA complexes studied by single-molecule
fluorescence microscopy
Stratmann, Sarah
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2017
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Stratmann, S. (2017). Assembly dynamics of supramolecular protein-DNA complexes studied by single-molecule fluorescence microscopy. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Chapter 3
Single-molecule studies of DnaB loading and
dynamics at the Escherichia coli replication fork
Sarah Stratmann, Zhi-Qiang Xu, Nick Dixon, Antoine van Oijen
(Manuscript in preparation)
Loading of ring-shaped, replicative helicases onto DNA is a key step in the initiation of cellular replication. Helicases are motor proteins that catalyze unwinding of the DNA du-plex and that coordinate leading and lagging-strand priming and polymerization steps. Using fluorescence-based single-molecule imaging we visualized individual E. coli DnaB hexameric helicases loading onto replication forks in a manner dependent on the pres-ence of the helicase loader DnaC. We observed DnaB stably associated with the replisome during replication elongation and show that exchange takes place much slower than the timescales associated with the enzymatic events responsible for lagging-strand synthesis, illustrating the importance of DnaB ensuring overall fork stability and integrity during genome duplication.
3.1 Introduction
Replicative helicases play a key role in genome duplication. Usually assembling into homo-hexameric (DnaB family, RecA-like ATPases (1)) or heterohomo-hexameric rings (MCM family, AAA+ superfamily ATPases (2)), helicases encircle single-stranded DNA to processively un-wind the parental duplex DNA in an ATPase-dependent manner, thereby providing the sin-gle-stranded DNA templates for DNA polymerase and primase activity (3, 4).
DNA replication in E. coli is initiated at a specific site on the genome, the origin of replication (oriC) (5, 6), and occasionally re-initiated when the replisome collapses, particularly at DNA lesions (7). During initiation, loading of the helicase DnaB onto forked DNA templates is the key determinant for replisome assembly. The hexameric helicase provides interaction sites for primase molecules that cooperatively produce the RNA primers required for initiation of DNA polymerase activity (8, 9). The polymerase III holoenzyme (Pol III HE), with multiple copies of the polymerase III core enzyme being present at the fork, is responsible for the syn-thesis of new DNA on the leading and lagging strand. The central organizing unit of the Pol III HE is the clamp loader, a heteropentameric protein that interacts with DnaB via the clamp loader τ subunits (10). Due to the weak and transient nature of the interactions between DnaB on the one hand and the primase and Pol III HE on the other, we have little knowledge about the nature of the interactions that allow the E. coli replisome to assemble, hold it together, and mediate its various enzymatic activities. However, the central role of DnaB in orchestrating the formation of the multi-enzyme replisomal complex at the replication fork is generally acknowledged (11). Consequently, the stability of the replicative helicase at the fork is likely to be an important parameter for the molecular mechanisms underlying replication initiation and processive fork progression, which has been substantiated only to a limited extend with experimental data on the kinetics of DnaB loading and its stability at the replication fork. Extensive studies on DnaB-family helicases revealed that helicase loading factors, in E. coli the DnaC protein, are required to position the toroidal hexameric helicase onto single-stranded DNA. The loading factor DnaC, belonging to the AAA+ superfamily of ATPases, forms a tight complex with the helicase, with the DnaB hexamer interacting with six units of DnaC at sub-nanomolar affinity in solution (12-14). Atomic-resolution structures of the DnaB-DnaC complex of a homologous bacterial system (15), as well as cryo-EM, SAXS (16) and hydrogen/ deuterium exchange studies (17) of the E. coli DnaB-DnaC complex suggest that ATP-bound DnaC destabilizes the DnaB hexamer to a three-tiered, right-handed, cracked-ring structure and enables deposition onto single-stranded DNA (18). Subsequent ATP hydrolysis by DnaC relieves the DnaB-DnaC interaction, triggering dissociation of DnaC and activation of DnaB (14). The cellular stoichiometry of the DnaB-DnaC complex required for loading might be lower with regard to DnaC and more diverse (19). It is also unclear whether DnaC is able to remain bound to DnaB during unwinding and replication. High expression levels of DnaC result in DnaB inhibition and replisome stalling, suggesting that DnaC interacts with the ac-tively unwinding DnaB at the fork (12, 20).
The mechanism by which DnaB utilizes the free energy from ATP hydrolysis to unwind DNA is not clear. Key to understanding this mechanochemical coupling is the visualization of the conformation of the DnaB subunits during unwinding. A crystal structure of a staircase-like conformation of the homologous Bacillus stearothermophilus DnaB hexamer bound to
sin-Introduction
gle-stranded DNA (21) suggests a hand-over-hand movement of subunits which underlies duplex destabilization. The intricate nature of the coupling between unwinding and the ac-tivities of the other replisomal components becomes clear when comparing DnaB velocities outside and within the replisome. On its own, DnaB is reported to unwind DNA at a rate of 35 bp/s (10) to 291 bp/s (19). The latter study demonstrated a low processivity of a few tens of base pairs due to a high dissociation rate of 10-20 s-1. A recent single-molecule, mag-netic-tweezer study on DnaB unwinding showed that DnaB pauses frequently between phases of unwinding activity, with unwinding rates of 50 bp/s, reconciling the velocities measured in bulk on different DNA substrate lengths (22). This study also reported much higher proces-sivities of up to 1 kbp. Within the replisome, DnaB unwinds DNA at a rate of about 400 bp/s (10). The increase in velocity might relate to the interaction with the τ subunits of the clamp loader complex (10). Dilution experiments demonstrated that DnaB is stably associated with the replisome and highly processive, together with with the leading and lagging-strand poly-merases, supporting a model wherein the DnaB in the context of the replisome is stabilized by protein-protein interactions with the Pol III HE (10).
The stability of DnaB at the replication fork has also been a point of interest in the context of replisome pausing, events that recently were shown to occur predominantly as a result of the replisome colliding with noncovalently bound protein complexes on genomic DNA (23). After pausing, replication can resume if the replisome is still intact or fork regression and re-loading is needed in the case of replisome collapse. Such replisome rere-loading requires several cofactors and recombination enzymes, representing a challenge to genome stability (24, 25). Different in vitro studies have reported stabilities of blocked replisomes of 5 mins (26) to 60 mins (27), concluding that at least the replicative helicase might stay bound at the fork to pro-vide a time window for replisome assembly without need for fork regression and activation of reloading pathways.
Most of our understanding of the bacterial replisome is based on ensemble-averaging bio-chemical experiments that played a key role in the elucidation of enzymatic mechanisms, assembly pathways, and protein interaction networks. However, it is challenging in bulk as-says to visualize the rare or intermediate events that are critical to the dynamic pathways and events at the replication fork. Recent advances in single-molecule methods have enabled the real-time visualization of the activity of individual replisomes and have provided new insights into the molecular mechanisms underlying the replication reaction. Challenging the canoni-cal textbook view of a stable replisome that recycles the polymerases to support the synthesis of a large number of Okazaki fragments, single-molecule methods have been used to show that at physiological protein concentrations, replicative polymerases are exchanged in and out of the replisome at a very high rate (Lewis, Spenkelink et al., unpublished results), (28, 29). These single-molecule studies show that multiple weak protein contacts within the replisome enable exchange when performed in the presence of competing proteins in solution, while replisomal components remain stably bound in the absence of competing factors (30). While bulk dilution experiments on pre-assembled replisomes demonstrate the theoretical stabil-ity of the multi-protein complex, the single-molecule experiments visualizing exchange in the presence of physiologically relevant protein concentrations suggest great plasticity of the replisome in replacing parts.
In light of these studies on replisomal dynamics and plasticity, the stability of DnaB appears to be a critical parameter for replisome integrity. If so, is indeed a single helicase supporting pro-cessive replication? Does DnaB spontaneously disassemble from the fork during replication initiation or progression? To address these questions, we present here a single-molecule as-say to directly visualize individual DnaB helicase complexes during replication initiation and replisome assembly. We show that DnaC-mediated loading of DnaB results in a long-lived state on single-stranded DNA that allows subsequent replisome assembly. Single helicases are able to processively unwind DNA in the context of replication and are not regularly exchang-ing in the presence of excess DnaB. Our results illustrate the importance of the replicative helicase at the core of the replisome. While the enzymes acting on the leading and lagging strand such as polymerases, clamp loader complexes and primases frequently dissociate and exchange with proteins in solution, DnaB ensures replisome integrity by providing a stable interaction platform.
3.2 Results
3.2.1 DnaB – association kinetics at forked DNA
Previous studies have established fluorescence methods for single-molecule resolved, in vitro replication reactions based on a M13 rolling circle substrate (31). Here, coupled leading and lagging strand replication in the presence of purified replisomal proteins is observed by in-troducing a dye (SYTOX Orange) during the reaction that stains exclusively double stranded (ds) DNA products. Existing protocols for E. coli replication showed that replisomes can be pre-assembled on forked DNA in the presence of DnaB and Pol III HE (32, 33). In order to subdivide and quantify efficiencies of DnaB loading and replication initiation, we adapted the pre-assembly conditions and loaded only DnaB in the presence of its loader protein DnaC and ATP onto surface-immobilized forked M13 before introducing the remaining replication pro-teins. We used a double-stranded M13 rolling circle template (34) with a 90 nt single-stranded (ss) fork providing a sufficient platform for helicase loading (Fig. 1A). For site-specific fluo-rescent labeling of the naturally cysteine-less DnaB, we engineered a mutant with a cysteine at the linker region between the N-terminal and C-terminal domain that does not interfere with the enzymatic function or known protein interaction sites (Suppl. Fig. 1). We immobilized the forked DNA by coupling to a functionalized coverslip surface of a microfluidic flow channel. By using SYTOX, the positions of the DNA molecules were localized with total-internal-re-flection microscopy. When injecting E. coli DnaB-Alexa647 (7.5 nM hexamer) and wild-type DnaC (45 nM) into the flow channel, we observed a stepwise increase in DnaB intensity, co-localizing with M13 (Fig. 1B, Suppl. Fig. 2A). Highly efficient, specific fork binding by DnaB proved to be dependent on DnaC and ATP (Suppl. Fig. 2C).
By applying a change-point algorithm, we tracked the intensity steps between the loading events as well as the corresponding dwell times (Fig. 1 C, D). By comparing the uniform distribution of intensity steps during M13 fork loading with equilibrium loading onto a 58 nt oligonucleotide coupled to the surface and taking the footprint of DnaB of 20 +/- 3 nt per hexamer (35) into account, we interpret the observed steps as loading events of single helicases, in agreement with the expectation that association of DnaB onto ssDNA involves
Results
well-defined hexamers (18). The distribution of dwell times between loading events were fit-ted to an exponential, excluding the potentially undersampled short dwell times (< 9 s), giving an association rate constant kon(obs) of (0.059 +/- 0.003) s-1. Next to 168 steps of increasing intensity, we observed only 12 steps of decreasing intensity with varying step sizes, which we account for as either dissociation or photobleaching (Suppl. Fig. 2A, B). Therefore we inter-pret the loading reaction of the DnaB/C complex as robust over at least the measurement time of about 2 minutes. DnaC/ATP 5‘ DnaB DnaC/ADP dsM13 7.2 kbp A B C D Time (s) 0 15 30 45 60 75 4 8 Fluorescence intensity (e+04) 12 M13-SYTOX DnaB-Al647 Replisome assembly τ TIRF illumination (biotin-)PEG Association/ loading 0 1e-05 2e-05
3e-05 Loading steps M13 Equilibrium loading 58-nt oligo
Density
0 5 10 15
Fluorescence intensity (e+04) *** kon(obs)=0.059 s-1 Dwell time t (s) 20 40 60 80 0 0 0.25 0.50 0.75 1.00 Cumulative density
Figure 1: Single-molecule DnaB loading reaction. A) Illustration of the single-molecule helicase loading assay.
Filled-in M13 is coupled via its 5’-tail to the functionalized surface of a microfluidic flow channel and stained with SYTOX Orange. DnaC, Alexa-dye labeled DnaB, and nucleotides required for loading are introduced into the flow cell. By adding the remaining replisome components, the replication reactions start and both DNA elongation as well as DnaB movement are tracked. B) Loading is observed as stepwise association of DnaB molecules at the fork. Intensity steps identified using a change-point algorithm; the corresponding fit is shown in red. C) The Gaussian intensity distribution corresponding to the loading steps is compared to an equilibrium loading reaction of DnaB onto an immobilized 58-nt oligonucleotide to determine the intensity corresponding to a single DnaB helicase. D) The cumulative distribution of dwell times between loading reactions is fitted to an exponential, providing a rate constant kon(obs) of (0.059 +/- 0.003) s-1 (at a concentration of (7.5 +/- 2.1) nM DnaB6), corresponding to a bimolecular
association rate constant of (7.9 . 106 +/- 2.1 . 106) M-1s-1. Undersampling of short dwell times was avoided by fitting
3.2.2 Loading of multiple helicases at the replication fork
After our observation of DnaB loading at the single-molecule level, we set out to determine the number of helicase hexamers loaded at the fork and the number associated with actively replicating replisomes. We first injected DnaB-Alexa647/DnaC complexes into the flow cell to allow association with the 5’ anchored, single-stranded DNA tails of the rolling-circle tem-plates and then applied the replication solution (Pol III HE, primase, SSB, dNTPs, NTPs). We also introduced SYTOX stain to allow the visualization of replication progression
simultane-A Fork velocity (kbp/s) C B 6 0 25 50 75 0 5 10 15 20 0.25 0.50 0.75 1.00 1.25 0 1 2 3 4 5
Time re. replication initiation (s)
Counts
DnaB-647 SYTOX merge
20 s 10 kbp
Arrival replication solution
# DnaB r = 0.11 0 50 150 250 350 0 2 4 -75 -50 -25 0 25 50 # DnaB # DnaB Loaded at M13 anchor Replicating 2 4 6 8 10 0 Time (s) Replication start (%) D E Arrival replication solution scale = 120 +/- 9 s 0 0.5 1 0 50 100 150 Time (s) 0.5 0.25 0.75 1
SYTOX (rel. int.)
# DnaB
Figure 2: Single-molecule replication initiation. A) Example two-color kymograph of replication initiation.
Intro-duction of the replication solution and SYTOX to surface-immobilized M13 molecules with pre-assembled DnaB-Alexa647 results in the formation of active replisomes. Replication is monitored by flow stretching the replicating rolling-circle templates and by fluorescence imaging of the labeled DnaB and SYTOX-stained dsDNA. B) Replica-tion is initiated within seconds after arrival of the replicaReplica-tion soluReplica-tion. Relative replicaReplica-tion start is given as percentage of the total amount of DnaB-loaded M13 forks. A Weibull distribution (in red) was used to estimate the character-istic dwell time (scale parameter) of replication initiation. C) Distribution of DnaB stoichiometries during loading onto the 90-nt fork and during replication at active forks. D) Trajectories of DnaB molecules (N=22) were aligned in time in reference to replication start and averaged (solid line). Initiation of replication results in a rapid reduction of the number of DnaB molecules present at the fork (colour coding as in C). E) No correlation is found between replication-fork velocity and the number of DnaB molecules per replisome (Pearson’s correlation coefficient r=0.11).
Results
ously with the distribution of DnaB along the DNA (Fig. 2A, movie 1). While continuously applying flow to stretch the double-stranded DNA product, we quantified progression of rep-lication by measuring the length of the SYTOX-stained DNA product and determined the copy number of DnaB at the fork by measuring the height of the DnaB fluorescence signal at the M13 tip (Suppl. Fig. 3).
Coupled replication could be observed within tens of seconds after introduction of the solu-tion containing the nucleotides and replicasolu-tion proteins, confirming that pre-loaded and stably-bound DnaB molecules at the fork are converted into active replisomes (Fig. 2B). By quantifying the DnaB fluorescence signal during pre-assembly of DnaB and by using the cal-ibration of the DnaB fluorescence signal as described in the previous section, we determined that 5.2 (25%-percentile=3.7, 75%-percentile=7.7) DnaB hexameric helicases were loaded at the 90-nt forked DNA substrate. This number is consistent with the reported footprint of DnaB of circa 20 nt (41) (Fig. 2C). After initiation of replication, however, the number of DnaB hexamers at the fork rapidly decreased (Fig. 2D), resulting in actively replicating com-plexes that contain 1.2 (25%-percentile=0.8, 75%-percentile=2.0) hexamers, as determined from aligning in time re replication start and averaging 22 DnaB fluorescence trajectories. We did not find a correlation between the replication rate and the number of hexamers in an active replisome (Fig. 2E). While more than one DnaB hexamer can be present in an active replisome, these additional helicases do not affect replication kinetics.
3.2.3 DnaB is stably integrated into the replisome during unwinding
To investigate the stability of DnaB at the replication fork, we followed DNA replication upon DnaB pre-assembly and compared the processivities of the replication reaction in the pres-ence and abspres-ence of DnaB/DnaC complexes in solution during the replication phase. Figure 3A shows no significant difference between processivities, demonstrating that the dissocia-tion rate of the originally loaded DnaB hexamer is slow enough not to represent an important role in determining processivity (38). To address the question whether replication velocity is affected by the presence of excess DnaB/DnaC in solution, we averaged trajectories (n=27) of replicating DnaB to visualize average replication progression as a function of time (Fig. 3B). These measurements do not show a significant change in the amount of replication product over time, whether excess DnaB is in solution or not. Combined with our observation that the processivities are also unaffected, we can infer that excess DnaB in solution does not pre-vent replisome stalling or collapse. Consistent with this notion is our observation that upon termination of replication, a large portion of stalled DNA forks (45%) still possessed DnaB (movie 2, Suppl. Fig. 4).
Next, we grouped multiple fluorescence trajectories of replicating DnaB molecules to ob-tain a time trajectory of the average DnaB signal at active forks (Fig. 3C). Irrespective of the presence of DnaB/DnaC in solution, the averaged fluorescence trajectories do not show sig-nificant signal deviations as a function of time, suggesting a stable integration of DnaB in the replisome. We occasionally observed stepwise changes in DnaB intensity at replication forks, which mostly corresponded to stalling of replication forks and subsequent increase in dynam-ics of DnaB loading (Fig. 3D). The majority of these newly-loaded helicases were observed to
bind to stalled forks and did not seem to re-inititate replication progression (Fig. 3E). These observations are consistent with the idea that stalling of a replisomes is associated with the production of single stranded DNA, with which DnaB can associate (42).
To further investigate the stability of DnaB in actively replication replisomes, we performed a Fluorescence Recovery after Photobleaching (FRAP) experiment. Here, we photobleached
DnaB-647 0 2 4 6 0 10 20 30 Time(s) # DnaB 5 µm 0 20 40 60 0 50 100 150 200 DnaB preassembly DnaB solution Counts Processivity (kbp) A B F 5 10 15 20 25 0 0.2 Velocity (kbp/s)
# DnaB binding at fork
0 50 100 200 Time (s) 150 0 2 #DnaB at fork 1 3 0 50 100 200 Time (s) 150 0 10 20 Replication product (kbp) D Active replisome Stalled replisome Non-replicated M13
High laser intensity SYTOX 20 s 10 kbp DnaB C E G
High laser intensity
0 30 60 120 Time (s) 90 0 5 10 15
# DnaB FRAP events
Figure 3: Stability of DnaB at the replication fork. A) Comparison of replication processivities in the absence and
presence of DnaB and DnaC during replication after pre-assembly onto forked M13. B) Fork progression in the ab-sence (N = 16) and preab-sence of DnaB in solution (N=13). Trajectories were averaged and the movement of the repli-cation fork plotted against time (color coding as in A). The shaded areas present the standard deviation of the average trajectory, obtained from the smoothed, single trajectories. The single trajectories are depicted in Suppl. Fig. 5. C) Number of DnaB hexamers at the fork during fork progression, corresponding to the trajectories in B). Fluorescence trajectories were smoothed and averaged over time. The single trajectories are depicted in Suppl. Fig. 5. D) Example kymograph and stoichiometry trajectory showing DnaB reloading at the fork. E) Reloading of DnaB at replication forks and the corresponding instantaneous fork velocities. F) Fluorescence Recovery After Photobleaching (FRAP) of DnaB during replication. DnaB reloading was quantified and classified according to whether they happened on unreplicated M13, active replicating forks or stalled forks (NDnaB before bleaching = 316, whereof bound to replication
prod-ucts NDnaB before bleaching, products = 76). G) Example kymograph of a bleached replisome with reappearance of DnaB signal
Discussion
the DnaB-Alexa647 fluorescence during replication by applying high laser intensity, followed by imaging of the replication reaction at reduced excitation intensity to monitor whether new, unbleached DnaB moleculaes are replacing the bleached ones (Fig. 3 F). Only little recovery is observed at (stalled) M13 sites and, in particular, in actively replicating forks. This observa-tion shows that the DnaB helicase is stably integrated in the replisome and does not frequently exchange during replication.
3.3 Discussion
With this study, we aimed at resolving the loading kinetics of the bacterial DnaB helicase and its stability during replication. Previous studies have inferred life times of several minutes of replisomes under dilution conditions (10), suggesting that DnaB molecules integrated into a functional replisome are exceedingly stable. On the other hand, the reported stability of DNA-bound DnaB in the absence of other replisomal components varies between millisec-onds (19) and secmillisec-onds (22). Considering the increasing amount of studies demonstrating that, at cellular protein concentrations, the replisome does not have a static composition, but rather exchanges its components frequently (28, 30), we asked whether the replicative helicase is also prone to exchange dynamics. Therefore we quantified fluorescence signals from sin-gle DnaB molecules to analyze the binding kinetics of DnaB during DnaC-mediated loading onto single stranded DNA as well as its association lifetime within active replisomes. By pre-assembling DnaB in a DnaC-mediated reaction onto surface-anchored forked DNA templates, we show that the helicase loads as a hexamer, resides stably at the fork, and sup-ports assembly of the replisome on the timescale of minutes. The stability of DnaB at forks is an important parameter, since experiments show that re-initiation after replisome pausing or collapse occurs frequently.
We further demonstrate that loading of multiple helicases does not impact replisome for-mation or activity (Fig. 2). In the cell, DnaB is assumed to actively unwind at the oriC to provide a template for DnaG-mediated priming (43). Therefore, it is not unlikely that more than one helicase loads onto single-stranded regions within the fork. We observe that de-spite the presence of multiple DnaB hexamers at the fork, the majority of active replisomes that show activity contain single helicases. Very frequently, though, we observe two or even more helicases present in active replisomes, without affecting the velocity of the replication complex (Fig. 2E). A recent in vivo fluorescence study on replisome stoichiometries (44) also showed that two DnaB helicases can be integrated into a single replisome, without apparent disturbance for replisome formation or progression. The authors of this study speculated that the fluorescence tag (Ypet) they utilized to label DnaB might trigger dimerization of helicase hexamers. Using an organic fluorophore to label DnaB, we prevent potential dimerization or aggregation scenarios. Combining both in vitro and in vivo data, we can infer that E. coli repli-somes display different stoichiometries of DnaB during initiation and elongation. The exact nature of the arrangement and positioning of multiple helicases at the fork region remains unclear, however. Potentially, additional helicases are residing in front of the replisomal heli-case translocating in 5’ to 3’ direction and aid in clearing the DNA of protein obstacles (45).
Recent studies have demonstrated a hitherto largely unappreciated plasticity of the composi-tion of the replisomal complex, with a focus on polymerase exchange during lagging-strand cycling (29, 46). In the E. coli replication system, it has been shown that not just the core polymerase (the αεθ subunits) exchanges, but rather the entire clamp loader complex itself exchanges frequently (28). Considering the interactions DnaB facilitates with other proteins in the bacterial replisome (47), one can think of the DnaB helicase as a multi-interaction plat-form orchestrate the exchange events controlling lagging-strand priming, primer hand-off as well as polymerase exchange. To directly assess the stability of DnaB within the replisome, we visualized and quantified DnaB and observe a stable integration of the protein during replication. This high stability of DnaB, in contrast with highly dynamic exchange of the other replisomal factors, argues for a revised model in which DnaB, as opposed to the polymerase-clamp loader complex, acts as a central hub of the replication complex.
We also observed that new DnaB molecules can be loaded onto forks that have stalled (Fig. 3D, F). It seems likely that at these forks DnaB is still translocating and unwinding the du-plex, thus providing stretches of single-stranded DNA regions for loading additional DnaB. Importantly, SSB, known to block helicase loading onto coated single stranded DNA (48), ap-parently does not entirely prevent this association reaction. Binding of SSB to ssDNA lengths less than 35 nt results in a much weaker association (49) and allows in a cellular context the binding of accessory helicases such as Rep to the fork region (50). Short stretches of ssDNA might therefore favor DnaB binding in our experimental conditions and result in DnaB ac-cumulation at forks.
In conclusion, by applying single-molecule imaging tools to an in vitro-reconstituted E. coli replication reaction, we demonstrate a high stability of integration of the replicative heli-case DnaB into the replisome and argue that such stability aids replisome integrity. Having established a fluorescence-based assay for visualizing DnaB, future work will need to focus on characterization of replisome collapse at termination sites, well-defined roadblocks and damage sites, which would enable a better understanding of the loading and unloading mech-anisms during these challenges to replisomal integrity.
3.4 Materials and Methods
DnaB-H201C purification
Purification of DnaB-H201C was done as described previously (51). Briefly, overexpression of DnaB/C, cloned into a pCE30 vector, was performed in MC1061 cells from the tandem phage lambda promoters pL and pR. Cells were grown at 30°C until the optical density at 600 nm reached 0.6, then the temperature was lifted to 42°C for 2.5 h for protein expression. Cells were harvested and resuspended in lysis buffer (50 mM Tris-HCl, pH 7.6, 10% sucrose, 100 mM NaCl, 2 mM DTT, 10 mM spermidine, 3 mM EDTA, 0.25 mg/ml lysozyme), stirred for 2 h at 4°C, heated for 6 minutes at 37°C, stirred for 1 h at 4°C, and centrifuged at 12000 g. Proteins were precipitated with ammonium sulfate (0.2 g/ml), and collected at 40000 g for 45 min. The precipitate was resuspended in DEAE buffer A (30 mM Tris-HCl, pH 7.6, 20%
glyc-Materials and Methods
erol, 10 mM MgCl2, 2 mM DTT, 100 µM ADP), dialyzed against buffer A overnight, loaded onto a DEAE column (50 ml), and eluted with a 25-350 mM NaCl gradient (DEAE buffer B with 1 M NaCl). DnaB-containing fractions were precipitated with ammonium sulfate (0.4 g/ ml), centrifuged and resuspended in gel-filtration buffer (30 mM Tris-HCl, pH 7.6, 20% glyc-erol, 5 mM MgCl2, 2 mM DTT, 100 µM ADP, 200 mM NaCl) and applied to a size-exclusion column (SEC200).
Labeling
Purified protein was dialysed into labeling buffer (30 mM Tris-HCl pH 7.8, 5 mM MgCl2, 100 mM NaCl, 100 µM ATP, 15 % glycerol, 0.05% tween, 0.5 mM TCEP), labeled with 20x molar excess AlexaFluor-647 maleimide dye (Invitrogen) overnight, cleaned via desalting columns (Pierce Zeba Column 2 ml) and size-exclusion chromatography (SEC200). The final labelling ratio amounted to 2.6 +/- 0.5 dyes per DnaB hexamer.
Unwinding assay
Unwinding activity by labeled DnaB was monitored according to (16). A forked substrate was prepared by annealing a 5’-Cy5 oligonucleotide (5’-Cy3-TACGTAACGAGCCTG-C(dT)25-3’) to a 3’-black-hole-quencher oligonucleotide (5’-(dT)25-GCAGGCTCGTTAC-GTA-BHQ2–3’). Unwinding reactions were performed with 250 nM DnaB6, 1.5 µM DnaC, 100 nM forked substrate, 200 nM capture DNA (5’-GCAGGCTCGTTACGTA-3’), 1 mM ATP in unwinding buffer (20 mM Tris-HCl pH 7.5, 10 mM MgOAc, 80 mM K-glutamate, 0.1 mg/ ml BSA, 1 mM DTT) at 37°C.
Single-molecule DnaB loading and replication reactions
Double-stranded, forked M13 substrates were produced and coupled to the surface of PDMS based microfluidic channels as described previously (52). Alternatively, 100 pM biotinylated oligonucleotide (table 1) was coupled to the surface in an analogous protocol. Fluorescent DnaB6/DnaC (7.5 nM/45 nM) was applied in replication buffer (20 mM Tris-HCl pH 7.5, 10 mM MgOAc, 80 mM K-glutamate, 0.1 mg/ml BSA, 1 mM DTT, 0.03% tween-20, 1 mM Trolox, 40 mM glucose, 250 nM glucose oxidase, 60 nM catalase), supplemented with 1mM ATP or nucleotide analogs.
In preassembly-only conditions, replication reactions were started by applying τ3δδ’χψ (10nM), αεθ (40 nM), β2 (30 nM), DnaG (300 nM), SSB (250 nM), NTPs (each 250 µM), dNTPs (each 50 µM), ATP (1 mM), SYTOX Orange (50 nM) after a short wash of the flow channel with replication buffer. Otherwise, DnaB and DnaC were added also during the rep-lication reaction (15 nM DnaB6, 90 nM DnaC).
Microscopy and data analysis
We used an objective-based TIRF system for single-molecule imaging. SYTOX-stained DNA molecules and fluorescent DnaB molecules were excited with 532 nm and 647 nm solid-state lasers (Coherent) at 25 and 36 Wcm-2, respectively, at a slightly elevated angle above total in-ternal reflection. The resulting fluorescent signal was collected through a 100x oil-immersion TIRF objective (Olympus, 1.49 NA) and recorded on an EM-CCD camera (Hamamatsu). Metavue software was used for image acquisition. Background correction, gaussian fitting of molecules, colocalization and tracking were performed in ImageJ. Data were analysed in R. The mean of the intensity distribution of loading steps of DnaB (Fig. 1C) was used as a cal-ibration parameter to quantify the number of DnaB molecules during replication. We used the changepoint package in R (40) to analyse sequential loading steps (Fig. 1, Suppl. Fig. 2). Fluorescence trajectories of DnaB-Alexa647 were smoothed with a local polynomial regres-sion fitting (in-built method in ggplot2, span=0.5, degree=1) (Fig. 3, Suppl. Fig. 5).
3.5 Supplementary data
Suppl. Table 1: DNA oligos used in this study
DNA oligo Sequence (5’-3’)
4 nt linker M13 Biotin-(T4)AATTCGTAATCATGGTCATAGCTGTTTCCT
90 nt linker M13 Biotin-(T90)AATTCGTAATCATGGTCATAGCTGTTTCCT
58 nt oligo AGGTCGCCGCCCTACGTAACGAGCCTGC(A30)-Biotin
bulk unwinding oligo sense Cy3-TACGTAACGAGCCTGC(T25) bulk unwinding oligo antisense (T25) -GCAGGCTCGTTACGTA-BHQ2 bulk unwinding capture GCAGGCTCGTTACGTA
Supplementary data
Suppl. Figure 1: Bulk activity of DnaB-H201C. A) Homology model of the E. coli DnaB hexamer (17), docked to six
DnaC molecules (blue). The linker domains between the N-terminal and C-terminal domains of the DnaB subunits are indicated in red, the mutated labeling position H201 is indicated in blue. B) Size-exclusion chromatography elution profiles of unlabeled and labeled DnaB-H201C. C) DnaB-mediated unwinding reaction of a forked DNA substrate. The reaction is dependent on the presence DnaC and ATP (16). D) Bulk replication reaction on M13. M13 rolling circle reactions were performed in a bulk assay and applied to an agarose gel. As a negative control, the replication-deficient mutant DnaB-R231C (53) was used instead of the WT or H201C.
5 10 15 Elution volume (ml) 0 0.5 1 DnaB-H201C-Alexa647DnaB-H201C Absorption (normalized) linker domain (red) A B C DnaB-H201 Direction of unwinding
DnaC hexamer (blue) DnaB hexamer
5 10 15 Time (min) 0.1 0.2 0.3 Unwound pr oduct DnaB-WT/DnaC, ATP DnaB-H201C-Alexa647/DnaC, ATP DnaB-WT, ATP DnaB-WT/DnaC, AMPPNP 16 bp Cy3 BHQ-2 3‘ 5‘ D M13 educt DnaB-WT neg. contr ol DnaB-R231CDnaB-H201CDnaB-H201-Cy5 Replication products
Fluorescence intensity (/e+04) 0 20 40 80 120 5 15 5 15 2 8 2 6 2 8 2 6 100 60 Time (s) 0 20 40 6080100120 0 20 40 6080100120 0 20 4060 80100120 Time (s) Time (s) 0 20 40 60 80100120 A B C 0.2 0.8 1.0 Relative fluorescence Alexa647 Time (s) 0 200 400 600 800 kbleach= 0.01 +/-0.002 s-1 0.00 0.25 0.50 0.75 1.00 nucleotide ATP AMP - ATP ATP
DnaC + + + - + fork (nt) 90 90 90 90 4 0.4 0.6 0 14 0 20 4060 80100120
Loading efficiency onto forked M13
Suppl. Figure 2: Single-molecule DnaB loading. A) Example fluorescence trajectories of DnaB during loading onto
M13. Steps were detected using the distributed change-point package in R; fits are shown in blue. B) DnaB-Alexa647 bleaching at a laser density of 36 mW/cm2, measured after loading onto 90-nt M13 forks. A fit with a
mono-expo-nential decay (red) indicates the photobleaching rate at laser excitation intensity used for the single-molecule DnaB loading experiments (36 Wcm-2). C) Loading onto anchored M13 forks requires ATP or a nucleotide analog
(AMP-PNP) and the presence of DnaC. Loading onto filled-in M13 is efficient on a fork with a single-stranded 90-nt 5’ tail, whereas a 4 nt does not support helicase loading.
DnaB-647 SYTOX merge Time (s) #DnaB 0 0.5 1 0 10 20 30 10 s 10 kbp 0.64 kbp/s 20 kbp 10 s 0.55 kbp/s merge SYTOX DnaB-647 0 0.5 0 10 Time (s) #DnaB
Suppl. Figure 3: Example two-color
kymographs of replication with flu-orescent DnaB after pre-assembly. Fluorescent signals obtained from tracking DnaB-Alexa647 are con-verted to DnaB stoichiometries. SY-TOX-stained growing dsDNA mol-ecules colocalize with DnaB at the DNA tip, indicating the formation of replisomes.
Supplementary data
A mer ge DnaB-647 0 0.5 0 10 20 10 kbp10 s 10 kbp 10 s #DnaB Time (s) DnaB-647 SYTOX DnaB-647 0.25 0 25 50 75 0 2.5 5.0 7.5DnaB occupied stalled forks empty stalled forks
#DnaB Counts B mer ge DnaB-647 SYTOX
Suppl. Figure 4: Replisome stalling. A) Example two-color kymograph of a stalled replisome after replication of 70
kbp. A DnaB molecule remains at the fork for at least seconds. B) DnaB occupancy at stalled replication forks. In 45% of stalled forks, DnaB molecules could be colocalized with the corresponding DNA tip. DnaB stoichiometries are comparable with those obtained during replication (Fig. 2 C).
0 50 100 150 200 0 10 20 30 Replication product (kbp) 40 0 10 20 30 40 0 2 4 #DnaB at fork 0 50 100 150 200 Time(s)
Pre-assembly only Pre-assembly + solution
A B 0 50 100 150 200 Time(s) 0 1 2 3 4 0 50 100 150 200
Suppl. Figure 5: Trajectories of DnaB movement and intensity during replication. A) Replication is initiated without
additional DnaB and DnaC in solution after preassembly. Trajectories with a moving averaged amount of DnaB >3 molecules are excluded from further analysis. The remaining trajectories (N=16 out of 25) of the DNA fork position (top) and the DnaB stoichiometry at the fork (bottom) are individually smoothed with a local polynomial regression fitting and the resulting fits are averaged (blue line). The grey shaded area represents the standard deviation of the av-eraged trajectory. B) Same as in A), but with DnaB and DnaC present in solution during replication (N=13 out of 25).
3.6 References
1. D. D. Leipe, L. Aravind, N. V. Grishin, E. V. Koonin, The bacterial replicative helicase DnaB evolved from a RecA duplication. Genome research 10, 5 (Jan, 2000).
2. A. F. Neuwald, L. Aravind, J. L. Spouge, E. V. Koonin, AAA+: A class of chaperone-like ATPases associ-ated with the assembly, operation, and disassembly of protein complexes. Genome research 9, 27 (Jan, 1999). 3. S. J. Benkovic, A. M. Valentine, F. Salinas, Replisome-mediated DNA replication. Annu Rev Biochem 70, 181 (2001).
4. M. O’Donnell, L. Langston, B. Stillman, Principles and concepts of DNA replication in bacteria, archaea, and eukarya. Cold Spring Harbor perspectives in biology 5, (Jul, 2013).
5. M. L. Mott, J. M. Berger, DNA replication initiation: mechanisms and regulation in bacteria. Nature re-views. Microbiology 5, 343 (May, 2007).
6. K. E. Duderstadt, K. Chuang, J. M. Berger, DNA stretching by bacterial initiators promotes replication origin opening. Nature 478, 209 (Oct 13, 2011).
7. K. J. Marians, Mechanisms of replication fork restart in Escherichia coli. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 359, 71 (Jan 29, 2004).
8. Y. B. Lu, P. V. Ratnakar, B. K. Mohanty, D. Bastia, Direct physical interaction between DnaG primase and DnaB helicase of Escherichia coli is necessary for optimal synthesis of primer RNA. Proc Natl Acad Sci U S A 93, 12902 (Nov 12, 1996).
9. J. E. Corn, P. J. Pease, G. L. Hura, J. M. Berger, Crosstalk between primase subunits can act to regulate primer synthesis in trans. Molecular Cell 20, 391 (Nov 11, 2005).
10. S. Kim, H. G. Dallmann, C. S. McHenry, K. J. Marians, Coupling of a replicative polymerase and helicase: a tau-DnaB interaction mediates rapid replication fork movement. Cell 84, 643 (Feb 23, 1996).
11. T. R. Beattie, R. Reyes-Lamothe, A Replisome’s journey through the bacterial chromosome. Frontiers in microbiology 6, 562 (2015).
12. E. Wahle, R. S. Lasken, A. Kornberg, The dnaB-dnaC replication protein complex of Escherichia coli. I. Formation and properties. The Journal of biological chemistry 264, 2463 (Feb 15, 1989).
13. B. A. Learn, S. J. Um, L. Huang, R. McMacken, Cryptic single-stranded-DNA binding activities of the phage lambda P and Escherichia coli DnaC replication initiation proteins facilitate the transfer of E. coli DnaB heli-case onto DNA. Proc Natl Acad Sci U S A 94, 1154 (Feb 18, 1997).
14. M. J. Davey, L. Fang, P. McInerney, R. E. Georgescu, M. O’Donnell, The DnaC helicase loader is a dual ATP/ADP switch protein. The EMBO journal 21, 3148 (Jun 17, 2002).
15. B. Liu, W. K. Eliason, T. A. Steitz, Structure of a helicase-helicase loader complex reveals insights into the mechanism of bacterial primosome assembly. Nature communications 4, 2495 (2013).
16. E. Arias-Palomo, V. L. O’Shea, I. V. Hood, J. M. Berger, The bacterial DnaC helicase loader is a DnaB ring breaker. Cell 153, 438 (Apr 11, 2013).
17. S. Chodavarapu, A. D. Jones, M. Feig, J. M. Kaguni, DnaC traps DnaB as an open ring and remodels the domain that binds primase. Nucleic Acids Res 44, 210 (Jan 8, 2016).
18. V. L. O’Shea, J. M. Berger, Loading strategies of ring-shaped nucleic acid translocases and helicases. Cur-rent opinion in structural biology 25, 16 (Apr, 2014).
19. R. Galletto, M. J. Jezewska, W. Bujalowski, Unzipping mechanism of the double-stranded DNA unwind-ing by a hexameric helicase: The effect of the 3 ‘ Arm and the stability of the dsDNA on the unwindunwind-ing activity of the Escherichia coli DnaB helicase. J. Mol. Biol. 343, 101 (Oct, 2004).
20. G. C. Allen, Jr., A. Kornberg, Fine balance in the regulation of DnaB helicase by DnaC protein in replica-tion in Escherichia coli. The Journal of biological chemistry 266, 22096 (Nov 25, 1991).
References
Adopts a Nonplanar Conformation during Translocation. Cell 151, 267 (Oct, 2012).
22. N. Ribeck, D. L. Kaplan, I. Bruck, O. A. Saleh, DnaB helicase activity is modulated by DNA geometry and force. Biophysical journal 99, 2170 (Oct 6, 2010).
23. M. K. Gupta et al., Protein-DNA complexes are the primary sources of replication fork pausing in Esch-erichia coli. Proc Natl Acad Sci U S A 110, 7252 (Apr 30, 2013).
24. C. P. Guy et al., Rep provides a second motor at the replisome to promote duplication of protein-bound DNA. Mol Cell 36, 654 (Nov 25, 2009).
25. S. Lambert, A. Watson, D. M. Sheedy, B. Martin, A. M. Carr, Gross chromosomal rearrangements and elevated recombination at an inducible site-specific replication fork barrier. Cell 121, 689 (Jun 3, 2005).
26. P. McGlynn, C. P. Guy, Replication forks blocked by protein-DNA complexes have limited stability in vitro. J Mol Biol 381, 249 (Aug 29, 2008).
27. R. T. Pomerantz, M. O’Donnell, Direct restart of a replication fork stalled by a head-on RNA polymerase. Science 327, 590 (Jan 29, 2010).
28. Q. Yuan, P. R. Dohrmann, M. D. Sutton, C. S. McHenry, DNA Polymerase III, but not Polymerase IV, Must be Bound to tau-Containing DnaX Complex to Enable Exchange into Replication Forks. The Journal of biolog-ical chemistry, (Apr 7, 2016).
29. H. J. Geertsema, A. W. Kulczyk, C. C. Richardson, A. M. van Oijen, Single-molecule studies of polymerase dynamics and stoichiometry at the bacteriophage T7 replication machinery. Proc Natl Acad Sci U S A 111, 4073 (Mar 18, 2014).
30. C. Aberg, K. E. Duderstadt, A. M. van Oijen, Stability versus exchange: a paradox in DNA replication. Nucleic Acids Res, (Apr 25, 2016).
31. J. B. Lee et al., DNA primase acts as a molecular brake in DNA replication. Nature 439, 621 (Feb, 2006). 32. N. A. Tanner et al., Real-time single-molecule observation of rolling-circle DNA replication. Nucleic Acids Research 37, 6 (Mar, 2009).
33. S. Ticau, L. J. Friedman, N. A. Ivica, J. Gelles, S. P. Bell, Single-molecule studies of origin licensing reveal mechanisms ensuring bidirectional helicase loading. Cell 161, 513 (Apr 23, 2015).
34. Z. Zhang et al., Assembly of the bacteriophage T4 primosome: single-molecule and ensemble studies. Proc Natl Acad Sci U S A 102, 3254 (Mar 1, 2005).
35. G. J. Wuite, S. B. Smith, M. Young, D. Keller, C. Bustamante, Single-molecule studies of the effect of tem-plate tension on T7 DNA polymerase activity. Nature 404, 103 (Mar 2, 2000).
36. N. Y. Yao, R. E. Georgescu, J. Finkelstein, M. E. O’Donnell, Single-molecule analysis reveals that the lag-ging strand increases replisome processivity but slows replication fork progression. Proc. Natl. Acad. Sci. U. S. A. 106, 13236 (Aug, 2009).
37. R. E. Georgescu, I. Kurth, M. E. O’Donnell, Single-molecule studies reveal the function of a third poly-merase in the replisome. Nat Struct Mol Biol 19, 113 (Jan, 2012).
38. N. A. Tanner et al., E. coli DNA replication in the absence of free beta clamps. The EMBO journal 30, 1830 (May 4, 2011).
39. N. A. Tanner, A. M. van Oijen, Visualizing DNA replication at the single-molecule level. Methods in enzymology 475, 259 (2010).
40. R. Killick, I. A. Eckley, changepoint: An R Package for Changepoint Analysis. 2014 58, 19 (2014-06-25, 2014).
41. W. Bujalowski, M. J. Jezewska, Interactions of Escherichia coli primary replicative helicase DnaB protein with single-stranded DNA. The nucleic acid does not wrap around the protein hexamer. Biochemistry 34, 8513 (Jul 11, 1995).
42. A. Blastyak, DNA replication: damage tolerance at the assembly line. Trends in biochemical sciences 39, 301 (Jul, 2014).
43. L. Fang, M. J. Davey, M. O’Donnell, Replisome assembly at oriC, the replication origin of E. coli, reveals an explanation for initiation sites outside an origin. Mol Cell 4, 541 (Oct, 1999).
44. R. Reyes-Lamothe, D. J. Sherratt, M. C. Leake, Stoichiometry and Architecture of Active DNA Replication Machinery in Escherichia coli. Science 328, 498 (Apr 23, 2010).
45. I. J. Finkelstein, E. C. Greene, Molecular traffic jams on DNA. Annual review of biophysics 42, 241 (2013). 46. D. E. Johnson, M. Takahashi, S. M. Hamdan, S. J. Lee, C. C. Richardson, Exchange of DNA polymerases at the replication fork of bacteriophage T7. Proc Natl Acad Sci U S A 104, 5312 (Mar 27, 2007).
47. P. M. Schaeffer, M. J. Headlam, N. E. Dixon, Protein--protein interactions in the eubacterial replisome. IUBMB life 57, 5 (Jan, 2005).
48. S. R. Wessel et al., PriC-mediated DNA replication restart requires PriC complex formation with the single-stranded DNA-binding protein. The Journal of biological chemistry 288, 17569 (Jun 14, 2013).
49. G. Krauss, H. Sindermann, U. Schomburg, G. Maass, Escherichia coli single-strand deoxyribonucleic acid binding protein: stability, specificity, and kinetics of complexes with oligonucleotides and deoxyribonucleic acid. Biochemistry 20, 5346 (Sep 1, 1981).
50. J. Atkinson, M. K. Gupta, P. McGlynn, Interaction of Rep and DnaB on DNA. Nucleic Acids Res 39, 1351 (Mar, 2011).
51. M. C. San Martin, N. P. Stamford, N. Dammerova, N. E. Dixon, J. M. Carazo, A structural model for the Escherichia coli DnaB helicase based on electron microscopy data. Journal of structural biology 114, 167 (May-Jun, 1995).
52. N. A. Tanner et al., Real-time single-molecule observation of rolling-circle DNA replication. Nucleic Acids Res 37, e27 (Mar, 2009).
