This protocol demonstrates performing a single-molecule assay for live visualization of DNA unwinding by CMG helicase. It describes (1) preparing a DNA substrate, (2) purifying fluorescently labeled Drosophila melanogaster CMG helicase, (3) preparing a microfluidic flow cell for total internal reflection fluorescence (TIRF) microscopy, and (4) the single-molecule DNA unwinding assay.
Faithful genome duplication is essential for preserving the genetic stability of dividing cells. DNA replication is carried out during the S phase by a dynamic complex of proteins termed the replisome. At the heart of the replisome is the CDC45-MCM2-7-GINS (CMG) helicase, which separates the two strands of the DNA double helix such that DNA polymerases can copy each strand. During genome duplication, replisomes must overcome a plethora of obstacles and challenges. Each of these threatens genome stability, as failure to replicate DNA completely and accurately can lead to mutations, diseases, or cell death. Therefore, it is of great interest to understand how CMG functions in the replisome during both normal replication and replication stress. Here, we describe a total internal reflection fluorescence (TIRF) microscopy assay using recombinant purified proteins, which allows for real-time visualization of surface-tethered stretched DNA molecules by individual CMG complexes. This assay provides a powerful platform to investigate CMG behavior at the single-molecule level, allowing helicase dynamics to be directly observed with real-time control over reaction conditions.
DNA replication is tightly regulated, as a cell must duplicate its genome accurately to prevent mutations, disease, or death. Eukaryotic DNA replication is carried out by the replisome complex, which unwinds parental DNA and uses single-stranded DNA (ssDNA) as a template to synthesize new DNA onto. In the G1 phase, catalytically inactive double hexamers of MCM2-7 are loaded onto double-stranded DNA (dsDNA) at replication origins1. In the S phase, MCM2-7 complexes are activated by the binding of CDC45 and GINS2 to form 11-subunit CMG complexes (CDC45, MCM2-7, GINS). Each CMG initiates DNA unwinding in opposite directions, forming the core unit that the replisome arranges itself around3.
Two decades ago the CMG helicase was first identified as an 11-subunit complex, essential for DNA replication4. Since then, our understanding of CMG has advanced considerably, from loading and activation5,6, to DNA unwinding and termination7. Traditional biochemical and structural biology techniques have been critical to many of these discoveries; however, these methods were often limited in their ability to study the more dynamic aspects of CMG. Single-molecule methods use physical manipulation of individual biomolecules to measure or visualize their activity one molecule at a time. This can be used to provide insight into the real-time dynamics of proteins that are often missed or undetectable by other techniques8,9.
Here, we describe a total internal reflection fluorescence (TIRF) microscopy assay to visualize DNA unwinding by CMG helicase in real time. Purified, fluorescently labeled CMG is loaded onto the free 3' end of long DNA containing a pre-made DNA fork structure. Linear DNA is stretched on a biotin-PEG coverslip in a microfluidic flow cell by tethering each end of the DNA sequentially to the surface. This approach allows for more uniform DNA tethering, which significantly reduces the variation that must be accounted for during data analysis. In the presence of ATP-γ-s, CMG is loaded onto the single-stranded DNA at the 3' end of the fork. ATP-γ-s is a slowly hydrolyzable ATP analog, which permits CMG binding onto DNA but not unwinding. Subsequent addition of ATP, along with purified, fluorescently tagged RPA, activates CMG and initiates extensive DNA unwinding. Visually, CMG translocates along the DNA, leaving a growing tract of RPA-bound ssDNA behind it. The untethered DNA end travels with CMG, forming a "tight ball" due to compaction caused by RPA binding. The flow cell design allows the buffer to be exchanged at any point during unwinding, giving great control during and over each experiment.
This protocol is divided into four methods, which can be performed independently of one another. Section 1 describes the preparation of a 20-kb linear forked DNA substrate for single-molecule assays. Section 2 outlines the purification and fluorescent labeling of Drosophila melanogaster CMG (DmCMG). Key information about the expression of DmCMG is included in the notes section. Section 3 covers the preparation of a microfluidic flow cell that can be used on a TIRF microscope. Section 4 describes how to carry out the single-molecule DNA unwinding assay.
1. Preparation of 20-kb linear forked DNA used in single-molecule assays (Figure 1)
Figure 1: Graphical representation of DNA substrate preparation. (A) The biotinylated DNA fork end is created by annealing two partially complementary oligonucleotides: biotinylated and non-biotinylated one. (B) The main dsDNA fragment (~20 kb) is generated by restriction digest of pGC261 plasmid with two enzymes to create a linear DNA with different overhangs at each end. (C) The digoxigenin duplex DNA end is obtained by a PCR reaction performed in the presence of digoxigenin-dUTP, followed by restriction digest. Please click here to view a larger version of this figure.
2. Purification of Drosophila melanogaster CMG (Figure 2)
Figure 2: Purification of Drosophila melanogaster CMG from 4 L of Hi Five cells. The proteins were resolved on 4%-12% Bis-Tris polyacrylamide gel under 200 V in the presence of MOPS buffer. The sample is shown at each stage of the purification (cell lysate – 2 µL, FLAG elution – 10 µL, after the first ion exchange column – 10 µL, and after labeling and the second ion exchange column – 1 µL. (A) Coomassie staining confirms the presence of all 11 subunits of the CMG complex before (10 µL), and after (1 µL) fluorescent labeling. (B) The labeling efficiency of the MCM3 subunit was validated by scanning for Cy5 with a fluorescent image analyzer using a long pass red (LPR) filter. Please click here to view a larger version of this figure.
NOTE: To prepare fluorescently labeled Drosophila melanogaster CMG, a TEV cleavage site (ENLYFQG) followed by four Gly residues was introduced downstream of the N-terminal FLAG Tag on the MCM3 subunit (in pFastBac1 vector)10. To express the complex, the baculovirus expression system was used. For initial transfection, Sf21 cells were used separately for each CMG subunit (P1 virus stage). To further amplify the viruses, Sf9 cells were used (P2 virus stage). Subsequently, Sf9 cell cultures (100 mL for each CMG subunit; 0.5 x 106 cells/mL) were infected with 0.5 mL P2 virus supplemented with 10% fetal calf serum (P3 virus stage). To express the entire CMG complex in 4 L of Hi Five cells (1 x 106 cells/mL), 200 mL of P3 viruses were used for each of the subunits. After harvesting the Hi Five cells expressing the CMG complex, the cell pellet can be flash frozen in liquid nitrogen and stored at -80°C. Perform the whole purification on ice or at 4 °C. The buffers can be prepared in advance, providing that the reducing agents (DTT or 2-Mercaptoethanol) and protease inhibitors (CAUTION) are added just before the use. Ensure that all buffers are precooled in advance, filtered, and degassed.
3. Preparation of the flow cell (Figure 3)
Figure 3: Graphical representation of the flow cell preparation. (A) Cut double-sided tape to match the size of the glass piece. Align the slide on top of the tape and mark the position of each hole with a needle. Using a razor blade, cut around each hold to create a channel. (B) Peel one side of the tape and stick the tape onto the glass piece. Ensure that both holes are inside the channel. Peel the second end of the tape and stick the biotin-PEG coverslip on top. (C) Insert the polyethylene tubing into each hole and seal the tubing in place with epoxy, sealing each glass piece to the coverslip too. (D) After using both channels, pull out the tubing and place the flow cell into a staining jar filled with acetone. After approximately 24 h, the epoxy and tape will have softened, and the layers of the flow cell can be peeled apart. The glass pieces can be recovered and stored in acetone to be reused indefinitely for making the next flow cell. Please click here to view a larger version of this figure.
4. Single molecule TIRF assay to visualize CMG-mediated DNA unwinding
Figure 4: DNA tethering to the surface. (A) When tethering DNA substrates with biotin at both ends, the distance between the two tethers can vary depending on how the ends contact the surface (i). By using digoxigenin at one end, the tethering of each end can be temporally separated for more consistent tether distances and more uniformly stretched DNA (ii). (B) Example field of view showing DNA tethered by both ends (digoxigenin-labeled) and stained with fluorescent intercalating nucleic acid stain. DNA, which is tethered by both ends, appears as a line, while DNA tethered by only one end appears as spots. Ideally, the DNA should be tethered as densely as possible without overlapping other DNA. The image is 512 x 512 pixels (pixel size = 154.6 nm). Please click here to view a larger version of this figure.
When CMG unwinds DNA, a characteristic RPA tract will grow over time (Figure 5). The 5' end of the unwound DNA is tethered to the surface; hence, it is seen as a linear stretch of RPA signal between the tether and the fork. The 3' end is not tethered and, therefore, moves with the fork and is observed as a compact EGFP-RPA signal. The position of the compacted unwound translocation strand corresponds approximately to the position of the replication fork, which moves together with LD655-CMG visualized via a 640 nm laser.
It is important to minimize damage to the DNA substrate, as damage such as single-stranded DNA nicks reduces the number of observable unwinding events, limiting the amount of data that can be collected (Figure 6).
Figure 5: Single-molecule DNA unwinding assay. The DNA substrate is tethered to a coverslip surface. Purified CMG labeled with LD655 is incubated with the DNA for 15 min in ATP-g-s. ATP and purified EGFP-labelled RPA are added, initiating extensive DNA unwinding by CMG. A cartoon schematic (left) and kymograph of representative data (right) are shown. Please click here to view a larger version of this figure.
Figure 6: DNA damage reduces assay throughput. (A) CMG cannot unwind DNA past a break in the DNA backbone (DNA nick). A nick on the leading strand template causes CMG to slide off the DNA, and both CMG and leading strand template are lost. A nick on the lagging strand template causes the lagging strand template to separate from the rest of the DNA, and each DNA piece retracts to their respective tether. This is illustrated by (i) cartoon schematics and (ii) kymographs of these events (ii). Representative data with (B) a minimally damaged DNA substrate versus (C) a more damaged DNA substrate at (i) 5 min, (ii) 15 min, and (iii) 60 min in a single field of view. The more damaged DNA substrate does not generate long tracts of unwinding, as the CMGs encounter nicks earlier on, despite similar levels of unwinding activity (similar density of growing RPA spots at 5 min, indicating similar CMG loading/unwinding efficiency). The field of view is 512 x 512 pixels (pixel size = 154.6 nm). 1% laser power (488 nm) imaging EGFP-RPA. Scale bar showing 10 µm. Please click here to view a larger version of this figure.
This assay provides a platform to observe and investigate the real-time dynamics of individual CMGs, both in isolation and in the context of the desired additional factors. However, as with many single-molecule fluorescence techniques, there are some common challenges that can require optimization to overcome. These usually relate to imaging fluorophores over long periods of time (photobleaching, brightness), DNA substrate preparation (DNA damage), the quality of the flow cell surface (background noise, non-specific interactions), or the quality of the purified protein preparation (nuclease contamination, labeling efficiency).
Each fluorophore varies in photostability and brightness, so it is important to choose an appropriate molecule. When imaging fluorescently labeled oligomeric proteins, like RPA, a lower laser power can be used as many fluorophores will be excited in close proximity, generating a visible signal. For imaging single fluorophores, for example, CMG labeled on a single subunit, a higher laser power is needed to observe the fluorophore clearly. Fluorophore lifetime can be extended by minimizing laser exposure, such as by reducing the frequency at which images are taken. Additionally, exciting a fluorophore generates reactive oxygen species (ROS), which can contribute to photobleaching. Including an oxygen scavenging system in the imaging buffer can extend the lifetime of fluorophores by eliminating ROS. However, some oxygen scavenging systems can affect pH12.
Regarding DNA substrate preparation, it is crucial to minimize DNA damage, such as nicks or single-stranded gaps. Excessive damage prevents extensive DNA unwinding, limiting how much data can be collected. Damage can arise from mechanical shearing, excessive heating, as a result of nuclease contamination, or ROS generated during imaging. Shearing can be minimized by handling the DNA sample with care by using wide-bore tips for pipetting, pipetting slowly, and avoiding flicking the sample. The effect of ROS can be minimized by either reducing laser exposure or including an oxygen scavenging system in the imaging buffer. After the preparation of the DNA substrate, it is possible to use commercial DNA repair kits to repair the damage before performing an unwinding reaction.
The efficiency of DNA unwinding also depends on the purity and activity of CMG. It is a good practice to assess sample purity after each purification step by SDS-PAGE electrophoresis to determine where optimization is necessary. If too many contaminants are observed after the final step, it may help to modify the salt gradient volumes used for the elution from CaptoHiRes Q (5/50) column. It is also highly important to remove any excess fluorescent peptide used for the protein labeling, as it can create undesirable background on the coverslip surface. It is also essential to avoid nuclease contamination, as this can degrade the DNA substrate. After an experiment, staining the remaining DNA with SYTOX orange can be a good way to check whether the DNA has been degraded significantly or not. A certain level of DNA damage is unavoidable over the course of an experiment, but significant damage often indicates problematic nuclease contamination.
The assay is also inherently limited by the resolution of diffraction-limited spots, requiring fluorescent proteins to be hundreds of base pairs away (if not more) to distinguish them as separate. This limits the detail in which CMG progression and interactions can be observed.
The number of unwinding events we observe for each analysis varies. For a successful experiment, we expect to see at least several RPA tracts of sufficient length per field of view 512×512 pixels (pixel size = 154.6 nm). Multiple fields of view can be imaged in the same experiment, allowing more data collection when necessary. The tracts do not need to be of the same length nor reach the end of the DNA to be useful. For example, the average tether distance for each experiment can be determined by measuring the length of SYTOX-stained DNA prior to adding CMG. This can be used to estimate how much DNA has been unwound for any RPA tract (as long as enough DNA is unwound to visibly move the fork) by converting distance from 'µm traveled' to 'kb unwound'.
CMG exhibits unwinding activity on a variety of DNA substrates, but it is essential to provide a free 3' DNA end on a polyT flap of at least 30 nt to accommodate the footprint of CMG10. Including multiple biotin moieties at the fork ensures robust surface tethering. The rest of the DNA substrate can be re-designed in a multitude of ways, such as to include different DNA sequences, lengths and chemical modifications. The conformation of the DNA can be altered by using different concentrations of magnesium acetate. At higher concentrations (≥10 mM) of magnesium acetate, the RPA-coated ssDNA filament is compacted, leading to the DNA being pulled taught by RPA binding during unwinding. This can be useful as it prevents the DNA from moving excessively, allowing the position of CMG and of unwinding progression to be more accurately measured. At low concentrations (~3 mM) of magnesium acetate, the RPA-ssDNA remains relaxed throughout.
The described single-molecule assay represents a platform that can be built upon and modified to investigate further aspects of DNA replication. During DNA replication, CMG acts as a core that the replisome and its components assemble around. Therefore, additional purified proteins can be added to this assay, including accessory factors like TIMELESS, TIPIN, and CLASPIN, to study their effect on CMG dynamics. These proteins have been shown to affect the rate of replication forks13, but it is not clear how they affect the rate of CMG unwinding. Therefore, it would be interesting to investigate how different replisome proteins affect CMG using this assay. Addition of DNA polymerases may give better insight into DNA replication beyond DNA unwinding alone, as described previously with yeast proteins14. Furthermore, purification of modified CMG may provide a better understanding of how certain mutations or post-translational modifications affect helicase activity15,16. Additionally, designing different DNA substrates can allow DNA unwinding by CMG to be studied under a variety of conditions mimicking replication stress17. These modifications include DNA obstacles9,18, inter-strand crosslinks19,20,21, and discontinuities in the DNA strands22.
The authors have nothing to disclose.
We thank Gheorghe Chistol for providing pGC261 plasmid and the Francis Crick Institute Chemical Biology Facility for peptide synthesis and labeling. This work was funded by The Francis Crick Institute, which receives core funding from Cancer Research UK, the UK Medical Research Council, and The Wellcome Trust (CC2133).
2-Mercaptoethanol | Sigma-Aldrich | M3148-25ML | |
Acrylamide / Bis Solution 40% | BioRad | 1610148 | |
Agarose UltraPure | Invitrogen | 16500-500 | Used to prepare Tris Buffer (pH 8) |
ÄKTA Pure | GE Healthcare | – | Used with Fraction Collector F9-C; protein purification system |
ANTI-FLAG M2 affinity gel | Sigma-Aldrich | A2220 | |
APS 10% | Sigma-Aldrich | A3678-25G | |
ATP (200 mM) | Sigma-Aldrich | A7699-5G | |
ATP-g-s (100 mM) | Sigma-Aldrich | 11162306001 | |
Biotin-PEG coverslip | N/A | N/A | Prepared as described: https://doi.org/10.1016/j.ymeth.2012.03.033 |
Biotinylated anti-digoxigenin antibody | Perkin Elmer | N/A | |
Blu Tack | Bostik | N/A | Adhesive putty |
Boric acid | Thermo Fisher Scientific | B/3800/53 | |
BSA; 33 mg/mL | SIGMA-ALDRICH | A3858-10G | Diluted from the stock |
CaCl2 | Sigma-Aldrich | C7902 | |
CaptoHiRes Q (5/50) | Cytiva | 29275878 | Connected to the AKTA protein purification system; high-resolution ion exchange chromatography column |
Casein (5% in water, 50 mg/mL) | Sigma-Aldrich | C4765-10ML | |
ChemiDoc system | Bio-Rad | N/A | |
Clips | N/A | N/A | |
Cutting board | N/A | N/A | |
Dialysis tubing (3.5 kDa) | Fisher Scientific | 11425859 | |
digoxigenin-11-dUTP | Roche | 11209256910 | |
DNA loading dye 6x | NEB | B7024S | |
dNTP 10 mM | NEB | N0447S | |
Double-sided tape (0.14 mm thick) | N/A | N/A | |
DTT | Fluorochem Limited | M02712-10G | |
DYKDDDDK peptide | N/A | N/A | Synthetised by chemical biology STP of The Francis Crick Institute |
EDTA | Thermo Fisher Scientific | D/0700/68 | |
EGFP-RPA | N/A | N/A | Purified as desribed in ‘Single Molecule Analysis’, Series Methods in Molecular Biology, Vol. 783 (2011), Peterman, E. J. G. and Wuite G. J. L. editors, Humana Press. Protein information: 2 mM, human, expressed in E. coli. |
EGTA | Sigma-Aldrich | 03779-10G | |
Epoxy – 5 minute | Devcon | 20845 | |
Fujifilm FLA-5000 | Fujifilm | FLA-5000 | Fluorescent image analyzer |
GelRed Nucleic Acid Stain; 10000x | Cambridge Bioscience | 41003-BT | Example of nucleic acid stain which is a safer alternative to ethidium bromide. |
Glass or quartz slide with holes for tubing | Dremel Model 395 | 1 mm thick slide, cut to 2.4 cm x 1cm, two holes drilled 1.4 mm apart. The holes should be drilled to different sizes to accommodate different tubing at each end. | |
Glycerol | Fisher Scientific | G/0650/17 | |
Glycine HCl, pH 3.5 | Sigma-Aldrich | G8898 | |
Hamilton syringe, 1000 series GASTIGHT, PTFE luer lock 1010TLL, PTFE Luer lock (with slots), volume 10 mL | Merck | 26211-U | |
HEPES pH 7.5 | Sigma-Aldrich | H4034 | |
I-CeuI | NEB | R0699L | |
KCl | Fisher Scientific | P/4280/53 | |
Magnesium acetate | Sigma-Aldrich | M2545 | |
Maxi GeBaFlex-tube Dialysis Kit; MWCO 14 kDa | Generon | D055 | |
Metal spatula | N/A | N/A | |
Metal tweezers | N/A | N/A | |
MgCl2 | Sigma-Aldrich | M2393 | |
Millex-GP (Syringe filters) 0.22 µm | Merck | SLGP033RS | |
NaCl | Sigma-Aldrich | S7653 | |
Needle | N/A | N/A | |
NotI-HF | NEB | R3189L | |
NuPAGE 4%–12% Bis-Tris Protein Gels | Thermo Fisher Scientific | 10247002; | |
NuPAGE MOPS SDS Running Buffer | Thermo Fisher Scientific | NP0001 | |
Objective heater | Okolab | N/A | |
PCR mix – 2x | Prepared from Phusion DNA polymerase (20 µL), 10 mM dNTPs (40 µL), 5x Phusion HF buffer (400 µL) and water (540 µL). | ||
Peptide NH2-CHHHHHHHHHHLPETGG-COOH, labelled with LD655-MAL | N/A | N/A | peptide NH2-CHHHHHHHHHHLPETGG-COOH, labeled with LD655-MAL (Lumidyne Technologies) on the cysteine residue, was synthetised and purified by the peptide chemistry STP of The Francis Crick Institute |
pGC261 plasmid | N/A | N/A | https://doi.org/10.1016/j.cell.2018.10.053 |
Phusion High-Fidelity DNA Polymerase | NEB | M0530L | |
Plastic tweezers | N/A | N/A | |
Polyethylene tubing PE20 (inner diameter 0.015”, outer diameter 0.043”) | Becton Dickinson | 427406 | |
Polyethylene tubing PE60 (inner diameter 0.03”, outer diameter 0.048”) | Becton Dickinson | 427416 | |
Poly-Prep Chromatography Columns; 10 ml and 20 ml | Bio-Rad | 7311550 | |
Potassium Glutamate (L-glutamic acid potassium salt monohydrate) | Sigma | G1501-1KG | |
Protease inhibitor cocktail (cOmplete, EDTA free) | Roche | 5056489001 | |
Pump 11 Elite Infusion/Withdrawal Programmable Single Syringe | Harvard Apparatus | 70-4504 | |
QIAquick PCR Purification Kit | QIAGEN | 28104 | |
Razor blade | VWR International Ltd | 233-0156 | |
rCutSmart Buffer; 10x | NEB | B6004S | |
Sodium acetate | Thermo Fisher Scientific | S/2120/53 | |
Sortase (pentamutant) | N/A | N/A | Purified based on: https://doi.org/10.1073/pnas.1101046108 |
Spin-X Centrifuge Tube Filters; 0.22 µm cellulose acetate | Fisher Scientific | 10310361 | |
Sterile scalpel | N/A | N/A | |
Steritop Vacuum Driven Disposable Filtration System; 0.22 um; PES; | Millipore | S2GPT05RE | |
Streptavidin (1 mg/mL) in 1x PBS buffer | Sigma | S4762-10MG | 20 µL aliquotes |
SYBR Gold | Thermo Fisher Scientific | S11494 | Highly sensitive nucleic acid stain we used to visualise DNA fork substrate. |
SYTOX orange; 5 µM in DMSO | Thermo Fisher Scientific | S11368 | |
T4 DNA ligase | NEB | M0202M | |
T4 DNA Ligase Reaction Buffer | NEB | B0202S | |
TEMED | Sigma-Aldrich | T9281-25ML | |
TEV protease (1 mg/mL); EZCut | Biovision | 7847-10000 | |
Tissue grinders, Dounce type (40 mL, Wheaton) | DWK Life Sciences | 432-1273 | |
Tris base | Sigma-Aldrich | T1503 | Used to prepare Tris Buffer (pH 8) |
Tris HCl | Sigma-Aldrich | T3253 | |
Tween-20 | Promega UK Ltd | H5152 |
.