Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2021 Jun 4;49(10):5470-5492.
doi: 10.1093/nar/gkab239.

Unravelling the mechanisms of Type 1A topoisomerases using single-molecule approaches

Dian Spakman  1 , Julia A M Bakx  1 , Andreas S Biebricher  1 , Erwin J G Peterman  1 , Gijs J L Wuite  1 , Graeme A King  2
Affiliations
Review

Unravelling the mechanisms of Type 1A topoisomerases using single-molecule approaches

Dian Spakman et al. Nucleic Acids Res. .

Abstract

Topoisomerases are essential enzymes that regulate DNA topology. Type 1A family topoisomerases are found in nearly all living organisms and are unique in that they require single-stranded (ss)DNA for activity. These enzymes are vital for maintaining supercoiling homeostasis and resolving DNA entanglements generated during DNA replication and repair. While the catalytic cycle of Type 1A topoisomerases has been long-known to involve an enzyme-bridged ssDNA gate that allows strand passage, a deeper mechanistic understanding of these enzymes has only recently begun to emerge. This knowledge has been greatly enhanced through the combination of biochemical studies and increasingly sophisticated single-molecule assays based on magnetic tweezers, optical tweezers, atomic force microscopy and Förster resonance energy transfer. In this review, we discuss how single-molecule assays have advanced our understanding of the gate opening dynamics and strand-passage mechanisms of Type 1A topoisomerases, as well as the interplay of Type 1A topoisomerases with partner proteins, such as RecQ-family helicases. We also highlight how these assays have shed new light on the likely functional roles of Type 1A topoisomerases in vivo and discuss recent developments in single-molecule technologies that could be applied to further enhance our understanding of these essential enzymes.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Overview of DNA topology and its interplay with genomic processes. (A) The influence of torsional stress on the structure of DNA. Upper left: In torsionally relaxed DNA, the linking number is equal to the twist (Tw), which is ∼1 turn/10.4 bp (referred to here as Twrelaxed). Upper right: Structure of positively supercoiled (+SC) DNA. At low tensions (typically below ∼3 pN), the torsional stress is stored only as writhe (Wr >0), through the formation of left-handed plectonemes. Lower: Structure of negatively supercoiled (-SC) DNA. The torsional stress can be stored as both twist (Tw < Twrelaxed) or writhe (Wr <0), even at low tensions (<1 pN). Changes in twist can lead to the formation of denatured, underwound structures (which often exhibit a left-handed form, e.g. L-DNA), whereas negative writhe yields right-handed plectonemes. (B) The Twin-Domain model for transcription. Due to the fact that genomic DNA is (locally) torsionally constrained (depicted here by grey blocks), translocation of the transcription machinery results in discrete domains of negatively supercoiled and positively supercoiled DNA behind and ahead of the transcription bubble, respectively. (C) Formation of precatenanes during replication. Progressive strand separation by the replication machinery leads to the accumulation of positive supercoils ahead of the replication fork, which can be relaxed by rotation of the replication fork (fork swiveling). This results in entwinement of the two daughter strands (in the form of precatenanes). (D) Generation and resolution of DNA entanglements during DNA repair. A dsDNA break can be repaired without crossover in a three-step process: (i) Homologous recombination results in the formation of a double Holliday junction; (ii) The double Holliday junction is converted into a hemicatenane via the concerted action of a TopoIII topoisomerase and a RecQ-family helicase (known together as the dissolvasome); and (iii) The hemicatenane is resolved by the dissolvasome into two separate dsDNA molecules.
Figure 2.
Figure 2.
Structure and catalytic cycle of Type 1A topoisomerases. (A) Crystal structures of EcTopoI (left, PBD 1MW8) (93) and EcTopoIII (right, PDB 1I7D) (92) in a closed conformation. The four domains of the toroidal fold (domains I-IV) are indicated. Both crystal structures show an ssDNA molecule (blue) bound to the ssDNA binding site of domain I, III and IV. The decatenation loop, which is present in EcTopoIII, but not in EcTopoI, is highlighted in green. (B) Crystal structure (left, PDB 4DDU) (80) and schematic (right) of TmRG, with the topoisomerase and helicase domains shown in grey and purple, respectively. The helicase domain consists of two subunits, H1 and H2, with the latter containing a latch domain (highlighted in orange). (C) Schematics of the closed and open conformational states of Type 1A topoisomerases, showing the four domains of the toroidal fold, along with the CTD (domain V). The schematic is based on the crystal structure of full-length EcTopoI (PDB 3PWT). (D) Proposed model of key steps of the catalytic cycle of Type 1A topoisomerases during relaxation of supercoiled DNA. 1. The enzyme binds, in a closed conformation, to a local region of ssDNA (G-segment, black) and cleaves the backbone (via a transesterification reaction) to form an enzyme-bridged ssDNA gate. The transesterification site is highlighted by the red dot. 2. The enzyme undergoes a conformational change, resulting in gate opening. 3. A second DNA strand (T-segment, purple) enters the central cavity of the enzyme via the gate. 4. The enzyme-bridged ssDNA gate closes. 5. The ssDNA backbone is re-ligated and the enzyme adopts an open conformation. This results in a change of ±1 Lk. 6. The T-segment is released from the cavity. 7. The enzyme can either unbind from the DNA (in a closed conformation) or undergo further catalytic cycles. Note that this catalytic cycle is also relevant for decatenation, except that the T-segment, in that case, would come from a different DNA molecule and is therefore likely further away from the G-segment.
Figure 3.
Figure 3.
Schematic overview of single-molecule techniques used to study Type 1A topoisomerases. (A) A typical magnetic tweezers assay. Left: A DNA molecule is tethered between a glass surface and a paramagnetic bead. A magnet placed above the sample chamber allows force to be applied to the bead (and thus the DNA). When using two horizontally-aligned magnets (as shown here), torsional stress can also be generated in the DNA by rotating the magnets (and therefore the bead). Right: Sample bright-field images showing the change in diffraction pattern of the bead as a function of magnet height, from which the DNA length can be determined. Right panel adapted from Figure 2D of (131). (B) A typical dual-trap optical tweezers assay. Left: A DNA molecule is tethered between two dielectric beads trapped by strongly focused near-infrared (NIR) laser beams. The DNA molecule can be extended by displacing one of the beads (Δx), resulting in an applied force. The force leads to a small deflection of the laser beam, which can then be measured by back-focal plane imaging on a position-sensitive detector (PSD) (yellow). Right: Top-down schematic view of the PSD indicating the deflection of the trapping laser beam (red) due to applied force. (C)Left: Schematic of a smFRET assay based on confocal imaging of a protein (purple) diffusing in solution. Here, changes in protein conformation are detected by monitoring the FRET efficiency (i.e. low or high FRET) between a donor and acceptor fluorophore that are covalently linked, respectively, to relevant structural domains within the protein. The FRET efficiency depends on the distance between the fluorophores. Right: Principle of FRET: When the distance between donor and acceptor is small, electronic excitation of the donor results in non-radiative energy transfer to the acceptor, followed by acceptor fluorescence (i.e. high FRET). (D) Left: Schematic illustration of an AFM set-up used for imaging biological samples, such as DNA. The sample is probed using a flexible cantilever connected to a small tip. Interactions between the tip and the surface are measured by directing a laser beam onto the rear face of the cantilever and detecting the position of the reflected light on a PSD (yellow). As the sample is scanned in the x,y-plane, the signal on the PSD is kept constant by adjusting the z-position of a piezo actuator (orange) using a feedback loop between the actuator and the PSD. Note that the actuator can be connected to either the cantilever or the surface. A three-dimensional image of the sample can be obtained by measuring the changes in actuator z-position. This can be used, for example, to directly visualize the topology of a negatively supercoiled DNA plasmid (right). Image adapted from Figure 1B of (155).
Figure 4.
Figure 4.
Overview of magnetic tweezers studies used to probe the mechanism of supercoil relaxation by bacterial Type 1A topoisomerases. (A) Schematic ‘hat-curve’ showing the measured DNA end-to-end length (in the absence of enzyme) as a function of bead rotations for three different tensions. Cartoons depict the DNA structure (either denatured, underwound or plectonemes) and show how the structure depends on both the direction of supercoiling and the applied force. The schematic is based on information provided in (15). (B) Schematic depicting how TmTopoI can stabilize the formation of denatured, underwound DNA (at the expense of plectonemes) in negatively supercoiled DNA (ΔLk <0) at intermediate (e.g. 0.9 pN) forces. (C) Schematic of a magnetic tweezers assay based on a positively supercoiled DNA molecule (ΔLk >0) containing either a bulge or a mismatch. The bulge and the mismatch each enable the binding of a single topoisomerase enzyme to the DNA. (D) (i) Representative traces showing the change in DNA end-to-end length resulting from relaxation of positively supercoiled DNA containing a mismatch of 12 bp by EcTopoI and TmTopoI at 2 and 1.5 pN, respectively. The step-wise behaviour due to discrete bursts of activity separated by pauses (lag times) is indicated. (ii) Histogram of the measured burst sizes for EcTopoI. A Gaussian fit reveals a mean burst size of 1.03 ± 0.1 Lk. Note that TmTopoI displayed a similar burst size (43). Adapted from Figure 5 of (43) (Copyright (2002) National Academy of Sciences, U.S.A.). (E)Lower: Plots showing the decrease in the DNA end-to-end length due to the introduction of negative supercoiling (via magnet rotations) and subsequent increase in DNA end-to-end length due to relaxation of the negative supercoiling by (i) EcTopoI (at 0.7 pN), (ii) EcTopoIII (at 0.7 pN) and (iii) StrepTopoI (at 0.5 pN). Examples of bursts and lag times are indicated. Upper: Schematic depicting the change in Lk associated with data in the lower panels, where dashed and solid blue lines represent magnet-induced supercoiling, and enzyme-induced relaxation, respectively. Lower panels adapted from Figure 2A and D of (56) and Figure 5A of (120), respectively. Upper panels produced using the information provided in (56) and (120). (F) Comparison of the primary domains of EcTopoI, EcTopoIII and StrepTopoI, showing domains I-IV and the CTD (domain V) in red and blue, respectively. The positively charged region in the CTD of StrepTopoI containing multiple lysine repeats is shown in green. (G) Schematic representation of the use of PIFE to detect enzyme binding to a 22-nt bulge (containing a Cy3 dye) within positively supercoiled DNA in a magnetic tweezers assay. Upon binding, an increase in fluorescence intensity (i.e. PIFE) is observed. (H) Representative traces of Cy3 fluorescence intensity over time before (left) and after (right) addition of EcTopoI, obtained using the assay described in panel G (at 1 pN). Adapted from Figure 2A and B of (98). (I) Histogram of the fluorescence intensities extracted from the traces shown in panel (H). Adapted from Figure 2A and B of (98).
Figure 5.
Figure 5.
Gate dynamics of EcTopoI and EcTopoIII probed using magnetic tweezers. (A)Upper: Schematic of a gapped DNA substrate used by Mills et al. (99) to directly measure the gate opening dynamics of EcTopoI and EcTopoIII, respectively. Lower: Schematic of the enzyme bound to the ssDNA gap in a closed, cleaved and open, cleaved conformation, respectively. The latter conformation results in an increase in the DNA end-to-end length (ΔExtension). (B) Representative extension-time traces for the substrate described in panel A at 16 pN in the presence of EcTopoI (upper) and EcTopoIII (lower). The ligated state (purple), closed, cleaved state (green) and open, cleaved state (blue) are indicated. (C) Plots showing the closing rates (kclose) for EcTopoI (upper) and EcTopoIII (lower) as a function of the applied force. An exponential decay function was fitted to the data (red lines), which yielded the lifetime of the open, cleaved state (1/kclose) in the absence of force. (D) Plots showing the influence of the magnesium concentration ([Mg2+]) on the re-ligation rates (kligation) for EcTopoI (upper) and EcTopoIII (lower). A linear function was fitted to the data for EcTopoI and an exponential growth function was fitted to the data for EcTopoIII (yellow lines). Panels B, C and D adapted from Figures 3B, 4C and D of (99), respectively.
Figure 6.
Figure 6.
Decatenation activity of EcTopoI and EcTopoIII probed using magnetic tweezers. (A) Schematic showing the relative extension of two parallel DNA molecules tethered between a single magnetic bead and the surface at constant tension as a function of magnet turns. The sharp decrease in bead height upon a half turn of the bead is highlighted as Δh. While only one direction of magnet rotations (positive) is shown here, a similar behaviour is observed when applying negative turns of the magnet. The schematic is based on the information provided in (134). Cartoons depict the substrates used by Terekhova et al. (59) to study the decatenation activity of EcTopoI and EcTopoIII. Here, two DNA molecules, each containing a 27-nt bulge, are entwined (braided) by rotating the magnetic bead by up to 30 to 35 turns. The initial crossover angle (βinitial) is indicated. The green arrows highlight the change in bead height associated with each substrate. (B) Representative traces showing the change in bead height during decatenation by EcTopoI (left) and EcTopoIII (right) using the assay described in panel A (for a 30–35-turn braid) at 2 pN. The traces exhibit step-like behaviour, corresponding to bursts of activity separated by pauses (lag times) that together determine the decatenation rate, as indicated. Dark green vertical arrows denote changes in bead height due to magnet-induced rotations of the bead. Adapted from Figure 1F and I of (59). (C) Bar plot showing the mean burst size and lag time for EcTopoI (grey) and EcTopoIII (white), respectively, during decatenation of a 30–35-turn braid at 2 pN. The burst sizes are defined in units of ΔCa, where Ca is the number of catenanes (i.e. turns) removed. Adapted from Figures 2A and 3A, respectively, of (59). (D) Bar plot showing the effect of the magnitude of βinitial on the mean lag time for EcTopoI and EcTopoIII measured during decatenation of a 30–35-turn braid at 2 pN. Grey and white bars represent small (24°) and large (47°) angles, respectively. Adapted from Figure 3B of (59). (E) Change in bead height as a function of time before and after braiding two DNA molecules, each containing a gap of 37 nt, by one turn in the presence of EcTopoI (upper) and EcTopoIII (lower). The red lines indicate magnet-induced rotations of the bead. Adapted from Supplementary Figure S3A and C of (119).
Figure 7.
Figure 7.
The interplay of eukaryotic Type 1A topoisomerases with partner proteins, as determined using magnetic tweezers, optical tweezers and AFM. (A) Schematic of a magnetic tweezers assay used by Kasaciunaite et al. (117) to probe the unwinding activity of Sgs1. Here, the DNA substrate contains a gap of 38 nt, along with a 40-nt ssDNA ‘flap’. In the presence of ATP, duplex unwinding by Sgs1 (initiated at the ds/ssDNA junction) will result in an increase in the DNA end-to-end length (Δx), which will be reversed upon re-winding. (B) Representative traces showing the change in the number of unwound bp (derived from changes in the DNA end-to-end length) for the substrate shown in panel A in the presence of (i) Sgs1, (ii) Sgs1 and ScTopoIII-Rmi1 and (iii) Sgs1, ScTopoIII-Rmi1 and yeast RPA (at forces in the range of 10–35 pN). Note that ATP was present in all cases. Gradual rewinding events are highlighted in orange shading, while rapid renaturation events are indicated by dark blue lines. Adapted from Figures 1B, 4B and E of (117), respectively. (C) Representative fluorescence images and corresponding schematics for a DNA molecule containing mixed ds/ss regions (tethered between optically-trapped beads) after incubation with (i) human BTRR (red) or (ii) human BTRR and RPA (red and green, respectively). BLM and RPA were labelled with mCherry and mStrawberry, respectively. Images adapted from Supplementary Figure S9B of (74). (D) Representative fluorescence images (left) and corresponding schematics (right) for DNA substrates containing mixed ds/ss regions (tethered between optically-trapped beads) after sequential incubation in channels containing: (i) human RPA-mStrawberry (red) and BTRR (containing BLM-mCherry); (ii) human PICH-eGFP (green); and (iii) human RPA-mStrawberry and BTRR (containing BLM-mCherry, blue). Note that in (i) and (iii) BTRR is excluded from the ssDNA by RPA. The final image and schematic (iv) show a composite of the above results. Images adapted from Figure 5B of (74). (E) Representative AFM images and corresponding schematics of a relaxed DNA plasmid prior to (i) and after (ii) incubation with human PICH and TRR (in the presence of ATP). The white arrows in (ii) indicate the locations of increased DNA height due to crossovers of the double helix (i.e. writhe). The measured height is indicated by the colored scale bar. The white scale bar represents 100 nm. Images adapted from Figure 2C of (75).
Figure 8.
Figure 8.
Overview of single-molecule studies of RG activity. (A) Schematic of a smFRET assay used by Del Toro Duany et al. (122–124) to study conformational changes of the helicase domain of RGs. Donor and acceptor dyes are covalently linked to H2 and H1 subunits, respectively. The FRET efficiency will increase upon closure of the cleft between H1 and H2. (B) Histograms of the measured FRET efficiencies (obtained using the assay described in panel A) based on the helicase domain of TmRG in (i) the absence of ATP (or ATP-analogues), (ii) the presence of AMP-PNP and (iii) the presence of ADP-MgFx. In all cases shown, a ds/ssDNA substrate was present. Adapted from Figure 6C of (122) with permission from the PCCP Owner Societies. (C)Upper: Schematic of the assay employed by Ogawa et al. (114,121) to measure DNA overwinding by StRG using fluorescently-labelled magnetic beads that can freely rotate. Lower: Sample fluorescence images showing clockwise rotation of fluorescently-labelled beads, obtained using the assay shown above (corresponding to DNA overwinding by StRG at 0.5 pN and 71°C, in the presence of ATP). Lower panel adapted from Figure 2A of (121). (D) Plot showing the mean of the maximal overwinding rates (orange circles) and of the average overwinding rates (blue triangles) of StRG as a function of temperature for DNA substrates containing zero, one, two and four 30-bp mismatches, respectively, determined using the assay described in panel C (121). Black dots represent individual maximal rates, whereas green crosses highlight conditions where the enzyme was unable to overwind the DNA. Adapted from Figure 4 of (114). (E) Interaction of SsoRG2 with (i) negatively supercoiled and (ii) positively supercoiled DNA, respectively, in low ATP concentrations at 45°C and 0.2 pN, studied using a magnetic tweezers assay similar to that shown in Figure 3A. Lower: Changes in the DNA end-to-end length as a function of time. SsoRG2 binding and unbinding from the DNA are indicated by green and yellow arrowheads, respectively. Successive strand passage events (+1 Lk) are indicated by black arrowheads. Adapted from Figure 1C and E of (88). Upper: Schematics depicting the change in Lk associated with the data in the lower panels (produced using the information provided in (88)). (F) Interaction of SsoRG2 with (i) negatively supercoiled and (ii) positively supercoiled DNA, respectively, in the presence of both AMP-PNP and ATP at 45°C and 0.2 pN, studied using a similar assay as in panel (E). Lower: Changes in DNA end-to-end length as a function of time. Orange and red arrows indicate AMP-PNP binding and subsequent ATP hydrolysis (i.e. conversion from ATP to ADP•Pi), respectively. Adapted from Figure 4A and C of (88). Upper: Schematics depicting the change in Lk associated with the data in the lower panels (produced using the information provided in (88)). (G) Proposed model for supercoil generation by RGs, based on the findings from single-molecule studies. 1. RG binds to (or, in the case of SsoRG2, forms) a strand-separated bubble, allowing the topoisomerase domain (grey) to cleave one strand (the G-segment, black), resulting in the formation of an enzyme-bridged ssDNA gate. 2. The enzyme undergoes a conformational change, leading to gate opening. 3. Upon ATP binding, the helicase domain (purple) switches to a closed conformation, which induces partial rewinding of the bubble. This facilitates T-segment (green) binding to the central cavity of the open topoisomerase domain. 4. Subsequent ATP hydrolysis results in opening of the helicase cleft and re-formation of the bubble. 5. The enzyme-bridged ssDNA gate closes, followed by re-ligation of the G-segment, resulting in a change of +1 Lk. 6. The topoisomerase domain adopts an open conformation in order to release the T-segment from the cavity. Note that this catalytic cycle is also relevant for the (ATP-dependent) strand-passage mechanism during negative supercoil relaxation for SsoRG2.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Fuller F.B. Decomposition of the linking number of a closed ribbon: a problem from molecular biology. Proc. Natl. Acad. Sci. USA. 1978; 75:3557–3561. - PMC - PubMed
    1. Bates A.D., Maxwell A.. DNA topology. 2005; Oxford University Press.
    1. Strick T.R., Allemand J.F., Bensimon D., Bensimon A., Croquette V.. The elasticity of a single supercoiled DNA molecule. Science. 1996; 271:1835–1837. - PubMed
    1. Strick T.R., Allemand J.F., Bensimon D., Croquette V.. Behavior of supercoiled DNA. Biophys. J. 1998; 74:2016–2028. - PMC - PubMed
    1. Strick T.R., Croquette V., Bensimon D.. Homologous pairing in stretched supercoiled DNA. Proc. Natl. Acad. Sci. USA. 1998; 95:10579–10583. - PMC - PubMed

Publication types