Research ArticleLIFE SCIENCES

A conserved ATP- and Scc2/4-dependent activity for cohesin in tethering DNA molecules

See allHide authors and affiliations

Science Advances  27 Nov 2019:
Vol. 5, no. 11, eaay6804
DOI: 10.1126/sciadv.aay6804
  • Fig. 1 Biochemical purification of yeast cohesin.

    (A) Purified cohesin tetramer containing Smc1, Smc3, Scc1, and Scc3 was analyzed by SDS–polyacrylamide gel electrophoresis (PAGE) followed by Coomassie blue staining. Western blot analysis shows the mobility of Smc1 and Scc1. (B) Top panel: Representative micrograph of a BS3-crosslinked cohesin sample observed in negative stain EM. Scale bar, 50 nm. Bottom panel: Class averages obtained with RELION. A set of the best ~5000 particles was used for this classification. The size of the circular mask is 450 Å. (C) Coomassie blue staining of the purified Scc2-Scc4 complex. (D) ATP hydrolysis by yeast cohesin and cohesin ATPase mutant Smc3-K38I with or without the Scc2-Scc4 complex.

  • Fig. 2 Analysis of yeast cohesin on DNA curtains.

    (A) Schematic representation of double-tethered DNA curtains used in the study. (B) Image of cohesin tagged with quantum dots (magenta) bound to λ-DNA stained with YOYO-1 (green). Scale bar, 10 μm. (C) Survival probability plots of cohesin in the presence of ATP, ADP, ATPγS, or no nucleotide. (D) Lifetimes of cohesin (fast phase and slow phase) in the presence or absence of Scc2-Scc4 and different ATP analogs. Error bars are 68% confidence intervals from bootstrapping. (E) Image of a pair of double-tethered DNA curtains bound by cohesin. DNA molecules are in green, and cohesin is in magenta. Diagrammatic representation is shown (left). (F) Time-lapse images of a pair of double-tethered DNA curtains bound by cohesin as they are tethered. DNA molecules are in green, and cohesin is in magenta. Diagrammatic representation is shown (top). Pairing events were observed frequently in the DNA curtains. An average of 5 to 10 events per DNA curtain was detected.

  • Fig. 3 Cohesin bridges DNA in an ATP- and Scc2-Scc4–dependent manner.

    (A) Schematic representation of FE curve for λ-DNA exhibiting the presence (right diagram and graph) and absence (left diagram and graph) of protein DNA bridges. Dotted line is fit to worm-like chain for naked DNA. (B) FE curves for λ-DNAs preincubated with 1 nM cohesin and 2.5 nM Scc2-Scc4 complex and 1 mM ATP (Cohesin + Scc2/4), 1 nM cohesin and 1 mM ATP (Cohesin), 2.5 nM Scc2-Scc4 complex and 1 mM ATP (Scc2/4), or 1 nM cohesin ATPase mutant and 2.5 nM Scc2-Scc4 complex and 1 mM ATP (CohesinK38I + Scc2/4). Schematic diagram of the experimental design. After capturing a single DNA molecule between two optically trapped beads, DNA was incubated in the presence of protein in a relaxed conformation (3-μm bead distance) for 30 s in 50 mM NaCl and then moved to a buffer channel with 125 mM NaCl for extension and measurements. Only incubation with 1 nM cohesin and 2.5 nM Scc2-Scc4 complex and 1 mM ATP (Cohesin + Scc2/4) showed DNA bridging rupture events. (C) FE curves in the presence of increasing ionic strength. High salt favors topologically constrained and permanent DNA bridges. (D) Schematic representation of the experimental design to test cohesin second DNA capture. After capture of λ-DNA between the two optically trapped beads, DNA is extended and incubated for 30 s in the protein channel. DNA is moved to a buffer channel and then relaxed (3-μm bead distance) and incubated for 30 s before reextension to test for DNA bridges (E). The extended DNA is then incubated in a relaxed position in the protein channel and then moved to buffer channel and extended to confirm that bridges can be formed when protein is loaded while DNA is relaxed (F). (E) λ-DNA incubated with 1 nM cohesin, 2.5 nM Scc2-Scc4 complex, and 1 mM ATP in an extended conformation and then moved to a buffer channel (125 mM NaCl) in the presence of 1 mM ATP (buffer only, dark blue) or 2.5 nM Scc2-Scc4 complex and 1 mM ATP (+Scc2/4, light blue). DNAs were reextended, and the FE curves were recorded. (F) The λ-DNA molecules in (E) were incubated in a relaxed position (3-μm bead distance) in the presence of 1 nM cohesin, 2.5 nM Scc2-Scc4 complex, and 1 mM ATP DNAs. DNAs were moved to a buffer-only channel (125 mM NaCl containing 1 mM ATP) and reextended. FE curves show the presence of DNA bridging rupture events.

  • Fig. 4 Cohesin and Scc2-Scc4 mediate intermolecular DNA bridges that slide on DNAs.

    (A) Schematic representation of the experimental design for the dual-trap optical tweezer to generate permanent intermolecular cohesin bridges. Two λ-DNA molecules are tethered between the two beads and incubated in a relaxed position (3-μm bead distance) in the presence or absence of protein in buffer containing 50 mM NaCl. The relaxed molecules are then moved to a different channel containing 300 mM NaCl and reextended. Imaging is done before incubations and after reextension in a buffer containing 300 mM NaCl and 50 nM SYTOX Orange to visualize DNA. (B) Two λ-DNA molecules were tethered and treated as described in (A) and incubated with either (i) 1 nM cohesin, 2.5 nM Scc2-Scc4, and no ATP (Cohesin + Scc2/4, left); (ii) 1 nM cohesin, 2.5 nM Scc2-Scc4, and 1 mM ATP (Cohesin + Scc2/4, middle); or (iii) 1 nM cohesin ATPase mutant K38I, 2.5 nM Scc2-Scc4, and 1 mM ATP (K38I + Scc2/4, right). Imaging was performed before incubation and after DNA reextension in a buffer containing 300 mM NaCl to minimize DNA entanglement and 50 nM SYTOX Orange to visualize DNA. Images from three independent experiments are shown. Three independent experiments are shown for each category. (C) Schematic representation of the experimental design to test for sliding of permanent cohesin bridges (top diagram). Following the formation of an intermolecular cohesin bridge (see fig. S8 for details in bridge formation protocol), beads 3 and 4 were moved together in the x axis to slide the bridge along DNA1. Images showing two representative sliding experiments are shown. Experiments were performed in a buffer containing 300 mM NaCl and 50 nM SYTOX Orange. Movies of the experiments are shown in movies S4 and S5. The experiment was performed three times, and sliding was observed in all cases. (D) Schematic representation of the experimental design to disrupt intermolecular cohesin bridges. Following the formation of an intermolecular cohesin bridge, bead 3 is moved down in the y axis until one of the DNA ends loses contact with the bead. Imaging was performed before and after the pull in a buffer containing 300 mM NaCl and 50 nM SYTOX Orange. Representative experiment is shown. A movie of the experiment is shown in movie S6.

  • Fig. 5 Cohesin does not compact linear DNA molecules stretched under low force.

    (A) FE curve for λ-DNA preincubated with 1 nM human cohesin and 1 mM ATP in 125 mM NaCl buffer (hCohesin). Dotted line is fit to worm-like chain model. After capturing a single DNA molecule between two optically trapped beads, DNA was incubated in the presence of protein in 50 mM NaCl buffer in a relaxed conformation (3-μm bead distance) for 30 s and then moved to the 125 mM NaCl buffer channel for extension and measurements. No evidence of DNA bridges was observed under this condition. (B) FE curve for λ-DNA preincubated with 1 nM human cohesin, 2.5 nM yeast Scc2-Scc4, and 1 mM ATP in 125 mM NaCl buffer (hCohesin + Scc2/4). Experimental procedure as in (A). FE curves exhibited multiple rupture events indicating the presence of reversible and permanent DNA bridges. (C) DNA compaction trace for λ-DNA molecule extended using a force of 1 pN (top). The DNA was tethered between two beads. One bead was clamped (fixed), while a 1-pN force was applied to the second bead to maintain the molecule extended. The DNA was then incubated in the presence of 1 nM condensin (1 mM ATP in 50 mM NaCl) (left, magenta trace). The FE curve for the λ-DNA full extension after incubation is shown (bottom). Additional examples can be found in fig. S10. (D) DNA compaction trace for λ-DNA molecule extended using a force of 1 pN (top) in the presence of 1 nM cohesin and 2.5 nM Scc2-Scc4 complex (1 mM ATP in 50 mM NaCl) (right, yellow trace). The distance between the beads was recorded over time. The FE curve for the λ-DNA full extension after incubation is shown (bottom). Additional examples can be found in fig. S10. (E) Kymograms of single-tethered λ-DNA stained with (YOYO-1) during the incubation with yeast cohesin and Scc2-Scc4 in the presence of ATP in 50 mM NaCl buffer at a flow rate of 20 μl/min. HF, high flow. The free end of DNA is marked with orange arrowheads. No compaction of single-tethered λ-DNAs was observed. (F) Kymograms of single-tethered λ-DNA stained with (YOYO-1) during the incubation with yeast cohesin and Scc2-Scc4 in the presence of ATP in 50 mM NaCl buffer at a flow rate of 10 μl/min. The conditions are as in (E) except for the reduced flow rate. Slow compaction of single-tethered λ-DNAs was observed over time (orange arrowheads mark the free end of DNA). (G) Kymograms of single-tethered λ-DNA stained with (YOYO-1) during the incubation with yeast cohesin and Scc2-Scc4 in the presence of ATP in 50 mM NaCl buffer at stopped flow. The free end of DNA is marked with orange arrowheads. The HF phase at the end of the experiment shows that the DNA was compacted during the stopped flow phase. Note that under stopped flow conditions, DNA molecules that diffuse laterally on the flow chip can transiently cross the field of view and also appear in a kymogram representation. Examples are marked with asterisks (*). These events bear no relevance for the interpretations of the assay.

  • Fig. 6 Tentative models for permanent cohesin bridges.

    (A) Schematic representation of expected behavior of intramolecular cohesin tethers from the previously proposed ring model. The model proposes that cohesin co-entraps two DNAs within its ring structure, i.e., both DNAs occupy one physical space within cohesin. From this model, it is expected that cohesin should be fully displaced from λ-DNA molecules when tethering in cis as force is applied to separate the beads. This is not what it was observed experimentally (Fig. 3C and fig. S5). (B) Schematic representation of expected behavior of intramolecular cohesin tethers from the subcompartment model. The subcompartment model is based on the assumption that DNAs are located in different physical compartments. The prediction from the model is that cohesin cannot be fully displaced from λ-DNA molecules when tethering them in cis. This is what we observed experimentally (Fig. 3C and fig. S5). (C) Proposed model for a single cohesin complex with at least three subcompartments (cohesin pretzel). In this model, sister DNAs occupy two different chambers (K1 and K2) of the K (kleisin) compartment formed between the SMC ATPase heads and the Scc1 subunit (36). Two possible conformations of SMC hinges are shown. Note that the experimental data are also compatible with the possibility that both DNAs jointly travel through the two chambers (K1 and K2) of the K compartment. (D) Schematic representation of previously proposed cohesin handcuffs models holding sister DNAs in different compartments of two separate complexes, which also fits with our experimental observations.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/11/eaay6804/DC1

    Fig. S1. Topological loading of yeast cohesin on plasmid DNA.

    Fig. S2. Analysis of yeast cohesin on DNA curtains.

    Fig. S3. Intramolecular cohesin bridging requires ATP.

    Fig. S4. Purification of budding yeast cohesin ATPase mutant.

    Fig. S5. Permanent cohesin bridges are not displaced by physical stretching of λ-DNA.

    Fig. S6. Cohesin does not capture two λ-DNAs in sequential steps.

    Fig. S7. DNA friction experiments confirm the presence of cohesin complexes on extended λ-DNA.

    Fig. S8. Generation of permanent cohesin bridges using a quadrupole-trap optical tweezer.

    Fig. S9. Purification of human cohesin and yeast condensin.

    Fig. S10. Budding yeast condensin, but not cohesin, compacts λ-DNA against 1 pN stretching force.

    Table S1. Mass spectrometry analysis of cohesin wild type and ATPase mutant (Smc3-K38I) tetramer complexes and the loader complex Scc2-Scc4.

    Table S2. Mass spectrometry analysis of cohesin ATPase mutant (Smc3-K38I) tetramer peptides showing peptides containing the K38I mutation for SMC3.

    Movies S1 to S3. Time-lapse videos showing cohesin tethering.

    Movies S4 and S5. Time-lapse videos showing sliding of intermolecular bridges in a quadruple-trap optical tweezer.

    Movie S6. Time-lapse video showing pulling on intermolecular bridges in a quadruple-trap optical tweezer.

  • Supplementary Materials

    The PDFset includes:

    • Fig. S1. Topological loading of yeast cohesin on plasmid DNA.
    • Fig. S2. Analysis of yeast cohesin on DNA curtains.
    • Fig. S3. Intramolecular cohesin bridging requires ATP.
    • Fig. S4. Purification of budding yeast cohesin ATPase mutant.
    • Fig. S5. Permanent cohesin bridges are not displaced by physical stretching of λ-DNA.
    • Fig. S6. Cohesin does not capture two λ-DNAs in sequential steps.
    • Fig. S7. DNA friction experiments confirm the presence of cohesin complexes on extended λ-DNA.
    • Fig. S8. Generation of permanent cohesin bridges using a quadrupole-trap optical tweezer.
    • Fig. S9. Purification of human cohesin and yeast condensin.
    • Fig. S10. Budding yeast condensin, but not cohesin, compacts λ-DNA against 1 pN stretching force.
    • Legends for tables S1 and S2
    • Legends for movies S1 to S6

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Table S1 (Microsoft Excel format). Mass spectrometry analysis of cohesin wild type and ATPase mutant (Smc3-K38I) tetramer complexes and the loader complex Scc2-Scc4.
    • Table S2 (Microsoft Excel format). Mass spectrometry analysis of cohesin ATPase mutant (Smc3-K38I) tetramer peptides showing peptides containing the K38I mutation for SMC3.
    • Movie S1 (.mp4 format). Time-lapse videos showing cohesin tethering.
    • Movie S2 (.mp4 format). Time-lapse videos showing cohesin tethering.
    • Movie S3 (.mp4 format). Time-lapse videos showing cohesin tethering.
    • Movie S4 (.mp4 format). Time-lapse videos showing sliding of intermolecular bridges in a quadruple-trap optical tweezer.
    • Movie S5 (.mp4 format). Time-lapse videos showing sliding of intermolecular bridges in a quadruple-trap optical tweezer.
    • Movie S6 (.mp4 format). Time-lapse video showing pulling on intermolecular bridges in a quadruple-trap optical tweezer.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article