RESEARCH ARTICLE SUMMARY◥

STRUCTURAL BIOLOGY

Structure and dynamics of the yeastSWR1-nucleosome complexOliver Willhoft*, Mohamed Ghoneim*, Chia-Liang Lin*, Eugene Y. D. Chua,Martin Wilkinson, Yuriy Chaban, Rafael Ayala, Elizabeth A. McCormack,Lorraine Ocloo, David S. Rueda†, Dale B. Wigley†

INTRODUCTION: Canonical nucleosomescontain two copies of each of four histone pro-teins: H2A, H2B, H3, and H4. However, var-iants of these histones can be inserted byadenosine triphosphate (ATP)–dependentchromatin-remodeling machines. The yeastSWR1 chromatin-remodeling complex, a mem-ber of the INO80 remodeler family, catalyzesthe exchange of H2A-H2B dimers for dimerscontaining Htz1 (H2A.Z in human) in an ATP-dependent manner. However, the mechanismby which SWR1 exchanges histones is poorlyunderstood. Despite having aDNA translocasesubunit similar to that in the INO80 complexthat slides nucleosomes, no net translocationof nucleosomes has been reported for SWR1.Consequently, the function of the ATPase activ-ity, which is required for histone exchange inSWR1, has remained enigmatic.

RATIONALE: To obtain sufficient quantitiesfor structural analysis, we generated the com-plete 14-subunit yeast SWR1 complex in in-sect cells. Binding of nucleosomes to SWR1 is

stabilized in the presence of an ATP analog(ADP•BeF3), which we used to prepare a com-plex with a canonical yeast H2A-containingnucleosome. Structural analysis was under-taken by cryo–electron microscopy (cryo-EM).We also used single-molecule FRET (smFRET)techniques to probe the dynamics of nucleo-somes bound to SWR1. Fluorescent probeswere positioned on the H2A histones and theend of the DNA to monitor changes in nu-cleosome dynamics upon binding of SWR1and ATP (or ATP analogs).

RESULTS:We determined the cryo-EM struc-ture of the SWR1-nucleosome complex at 3.6-Åresolution. The architecture of the complexshows how the SWR1 complex is assembledaround a heterohexameric core of the RuvBL1and RuvBL2 subunits. The Swr1 motor sub-unit binds at superhelical location 2 (SHL2), aposition it shares in common with other re-modelers but not with its most closely relatedcomplex, INO80, which binds at SHL6-SHL7.Binding of ATP or ADP•BeF3 to the SWR1-

nucleosome complex induces substantial un-wrapping of the DNA wrap. Conformationalchanges in the motor domains of the Swr1subunit drive a single–base pair translocationof the DNA wrap from the DNA entry site.The single–base pair DNA translocation ac-

companies conformation-al changes in the histonecore that begin to desta-bilize the histone dimerinterface. Using smFRETmethods, we further pro-bed these conformation-

al changes to show how an increase in thedynamics of the SWR1-bound nucleosomesis dependent on binding of ATP but nothydrolysis.

CONCLUSION: The cryo-EM structure of theSWR1 complex bound to a nucleosome revealsdetails of the intricate interactions betweencomponents of the SWR1 complex and its nu-cleosome substrate. Interactions between theSwr1 motor domains and the DNA wrap atSHL2 distort the DNA, causing a bulge withconcomitant translocation of the DNA by onebase pair, coupled to conformational changesof the histone core that likely destabilize thedimer interface. Furthermore, partial unwrap-ping of the DNA from the histone core takesplace upon binding of nucleosomes to theSWR1 complex. Single-molecule data monitorthis unwrappingand showhow thedynamics arealtered by ATP binding prior to hydrolysis.▪

RESEARCH

Willhoft et al., Science 362, 199 (2018) 12 October 2018 1 of 1

The list of author affiliations is available in the full article online.*These authors contributed equally to this work.†Corresponding author. Email: david.rueda@imperial.ac.uk(D.S.R.); d.wigley@imperial.ac.uk (D.B.W.)Cite this article as O. Willhoft et al., Science 362, eaat7716(2018). DOI: 10.1126/science.aat7716

Structure of the SWR1-nucleosome complex. (A) 3.6-Å SWR1-nucleosome map. (B) Binding of SWR1-ADP•BeF3 to the nucleosome inducesmultiple changes: (i) The DNA wrap is peeled away by ~2.5 turns, (ii) DNA is translocated by one base pair, and (iii) SHL2 is distorted as aconsequence of motor domain closure. These distortions are a precursor to histone exchange and can be monitored by smFRET.

ON OUR WEBSITE◥

Read the full articleat http://dx.doi.org/10.1126/science.aat7716..................................................

on March 28, 2020

http://science.sciencem

ag.org/D

ownloaded from

RESEARCH ARTICLE◥

STRUCTURAL BIOLOGY

Structure and dynamics of the yeastSWR1-nucleosome complexOliver Willhoft1*, Mohamed Ghoneim2,3*, Chia-Liang Lin1*†, Eugene Y. D. Chua1,Martin Wilkinson1, Yuriy Chaban1‡, Rafael Ayala1, Elizabeth A. McCormack1,Lorraine Ocloo1, David S. Rueda2,3§, Dale B. Wigley1§

The yeast SWR1 complex exchanges histone H2A in nucleosomes with Htz1 (H2A.Z in humans).The cryo–electron microscopy structure of the SWR1 complex bound to a nucleosome at3.6-angstrom resolution reveals details of the intricate interactions between componentsof the SWR1 complex and its nucleosome substrate. Interactions between the Swr1 motordomains and the DNAwrap at superhelical location 2 distort the DNA, causing a bulge withconcomitant translocation of the DNA by one base pair, coupled to conformational changesof the histone core. Furthermore, partial unwrapping of the DNA from the histone core takesplace upon binding of nucleosomes to SWR1 complex.The unwrapping, as monitored bysingle-molecule data, is stabilized and has its dynamics altered by adenosine triphosphatebinding but does not require hydrolysis.

The specific incorporation of histone var-iants into nucleosomes can trigger changesin chromatin structure (1) and the recruit-ment of cellular mediators (2). Histone var-iants can be inserted by histone chaperones

during chromatin deposition (3), or at a later stage,by adenosine triphosphate (ATP)–dependent chro-matin remodeling machines (4). Several variantsexist for canonical H2A and H3 histones, suchas the replication-independent yeast H3 vari-ant H3.3, which is present at actively transcribedgenes (5), or human H2A.X and H2A.Z, both ofwhich have essential roles in DNA repair (6).The yeast SWR1 chromatin-remodeling com-

plex is a member of the INO80 remodeler family(7). SWR1 drives the exchange ofH2A/H2B dimersfor Htz1/H2B in an ATP-dependent manner (8).In humans, the equivalent histone exchange re-action is catalyzed by the SRCAP (9) and TIP60(10) complexes, although the latter complex alsocatalyzes histone acetylation (10). However, themechanism by which all of these complexes ex-change histones has remained enigmatic. Despitehaving a superfamily 2 (SF2) helicase/translocase(Swr1 subunit) at its heart, and despite the de-pendence of histone exchange on the ATPaseactivity of this subunit (8), no net translocation ofnucleosomes has been reported. By contrast, the

related yeast INO80 remodeler [reported, albeitcontroversially, to catalyze the reverse exchangereactionby replacingHtz1/H2A.ZwithH2A (11–13)]appears to use a mechanism of limited trans-location at the nucleosomal entry/exit region inorder to exchange a histone dimer (14). Recentstructures of the INO80-nucleosome complexplace the ATPase motor at superhelical locationSHL6-SHL7 (15, 16), where it could potentially ex-pose the H2A(Z)-H2B dimer for histone exchangeby translocating DNA toward the nucleosomedyad. By contrast, the SWR1motor subunit (Swr1)binds at SHL2 (17), a position it shares in com-mon with other remodelers (18–20).

Structure of a complex of SWR1 with abound nucleosome

Biochemical data have shown that SWR1 prefersnucleosome substrates with overhangs on bothsides, with at least one of these needing to be

long [>60 base pairs (bp)] (17, 21). We thereforeused a canonical H2A-containing yeast nucleo-some with overhangs on both sides (113N25),in the presence of the ATP analog ADP•BeF3, toprepare a complex suitable for structure deter-mination by cryo-EM (Fig. 1 and figs. S1 to S3).As shown previously (21, 22), and in common

with the related INO80 complex (23), the SWR1complex contains a single heterohexamer ofthe RuvBL proteins (Fig. 1B) that serves as ascaffold around which many of the other sub-units are assembled (Movie 1). Each ATPase siteis occupied by an adenosine diphosphate (ADP)molecule. If there are any additional functions ofthe RuvBL hexamer, these do not require ATPhydrolysis because catalytically dead ATPase sub-units show wild-type histone exchange activity(fig. S4).Themotor subunit of INO80-family proteins is

characterized by a large insertion in the secondATPase domain (7) (Fig. 1A). As with the INO80motor subunit (23), the insert in Swr1 forms alarge extended structure that makes intimatecontacts with the RuvBL hexamer (Fig. 1, B to E).However, the structures of this insert are quitedifferent, although largely planar in both cases.One notable difference between the insert struc-tures is a two-helix extension that protrudes be-tween two RuvBL subunits in SWR1 and makescontact with the Arp6 subunit, in a region thatis an insert in the actin fold (Fig. 1E). A func-tion for this interaction, other than architectural,is unknown.Aswith the INO80-nucleosome complex (15, 16),

the N-terminal region of the SWR1 complex[termed Subcomplex 1 (SC1), comprising the N-terminal region of Swr1 and the actin, Arp4,Swc4, Swc7, Yaf9, and Bdf1 subunits] is presentbut disordered in our structure.

Interactions between the Swr1 ATPasedomains and the nucleosome

The nucleosome is bound in a cleft between twolobes that extend from the RuvBL core of thecomplex. Multiple contacts are made with thehistone core and DNA wrap. The Swr1 subunitATPase domains are located at the canonicalSHL2 position seen in the structures of mostother remodelers (19, 20) and consistent withSWR1-nucleosome footprinting studies (17). Thereis clear density for a bound nucleotide (ADP•BeF3) at the ATP-binding site (fig. S3) and, as aconsequence, the domains are in the “closed”conformation observed in the ATP-bound stateof other SF2 translocases (24) such as Chd1 (19)(fig. S5), placing themotor domains on the “track-ing” strand consistent with 3′-5′ translocationtoward the nucleosome dyad axis (Fig. 2, A andB). Although located at SHL2, the binding siteis displaced by 1 bp toward the dyad relative toChd1 and Snf2, resulting in a ~35° rotation of thecomplex around the bound DNA duplex towardthe other gyre of the DNA wrap (Movie 2). TheDNA of the bound nucleosome at the SHL2 sitecontacts the N-terminal ATPase domain (HD1),as observed in other remodelers (19, 20), but as aresult of the 1-bp shift, it now makes much more

RESEARCH

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 1 of 8

1Section of Structural Biology, Department of Medicine,Imperial College London, London SW7 2AZ, UK. 2SingleMolecule Imaging Group, MRC London Institute of MedicalSciences, London W12 0NN, UK. 3Molecular Virology,Department of Medicine, Imperial College London, LondonW12 0NN, UK.*These authors contributed equally to this work. †Present address:Department of Molecular and Cell Biology, University of Leicester,Leicester LE1 9HN, UK. ‡Present address: Electron Bio-ImagingCentre (eBIC), Diamond Light Source, Harwell Science andInnovation Campus, Didcot, Oxfordshire OX11 0DE, UK.§Corresponding author. Email: david.rueda@imperial.ac.uk(D.S.R.); d.wigley@imperial.ac.uk (D.B.W.)

Movie 1. Rotating overview of SWR1-nucleosomecomplex coordinates fitted into 3.6-Å cryo-EMenvelope. Subunits are colored as in Fig. 1B.

on March 28, 2020

http://science.sciencem

ag.org/D

ownloaded from

extensive contacts with the other gyre, sug-gesting a much tighter interaction on the DNA(Fig. 2C). Furthermore, across the second ATPasedomain (HD2), the bound DNA shows substan-tial distortion (Fig. 2, D and E, and fig. S5). Un-like DNA binding across the motor domains ofChd1, the DNA duplex makes a sharp kink, the

result of which is lifting of the DNA from thehistone surface to make a bulge in the wrap(fig. S5). The kink is a consequence of two ahelices being rammed against the duplex (Fig.2E). In order to create this bulge, the DNA hasalso been translocated by 1 bp (Movie 3). Thissuggests that the power stroke is ATP-binding,causing a translocation of 1 bp together with adistortion of the bound DNA. A step size of onebase per ATP has been deduced for other SF1and SF2 translocases (15, 25–27).The location of the Swr1 ATPase domains at

the SHL2 position (Fig. 2, A and B) is quite dif-ferent from that of the INO80 complex, whichbinds at SHL6-SHL7 (14–16). However, SWR1differs from INO80 in a number of importantways. Notably, a single complex of SWR1 is re-quired for activity (21), whereas INO80 functionsas a dimer (28). An INO80 dimer is able to slidenucleosomes along DNA, whereas a SWR1 mono-mer instead swaps histone dimers (21) and isunable to slide nucleosomes (17). Although yeastINO80 has been reported to also catalyze histoneexchange (11, 14), this activity is controversial(12, 13) and it is not known whether exchangeactivity requires INO80 dimers. Furthermore,although SWR1 is unable to slide nucleosomes,the ATPase activity of the “motor” domains isnonetheless essential for activity (8). Limited

sliding of DNA within the confines of the DNAwrap has recently been demonstrated for SWR1(29). The tighter interactions between the N-terminal motor domain (HD1) and the otherDNA gyre may act as a block to prevent nettranslocation by SWR1 but appears to allow

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 2 of 8

Movie 2. Rotation of Chd1 motor domainrelative to Swr1 motor. Chd1 (pink, PDB ID5O9G) is shown moving from a superpositionwith the Swr1 motor domain to its positionin the deposited Chd1-nucleosome structure.This shows that Swr1 is rotated by 35° aboutthe DNA axis.

Movie 3. Morph between canonical nucleosome(PDB ID 1ID3) and Swr1-bound nucleosome,which highlights the translocation event inducedby SWR1 binding as well as the distortion of thehistone octamer from the canonical positions.

Fig. 1. Overview of the SWR1-nucleosome complexarchitecture. (A) Linearizedcartoon of Swr1 subunit showingdomain boundaries. HD1 andHD2 are the two ATPase domains;HSA is the helicase/SANT-associated domain. (B) Overviewof SWR1-nucleosome complexat 3.6 Å (left) and builtcoordinates (right). (C) Viewof Swr1 insert (ribbon) insideRuvBL1/RuvBL2 ring structure(surface). (D) Rotated viewof SWR1, showing Swr1 insertfollowing the inner contourof the RuvBL1/RuvBL2 ring,with the Swr1 insert protrusionemerging from the ring.(E) Close-up of Swr1 protrusioninteracting with Arp6.

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

limited local translocation of the DNA wrapleading up to that block.

Other contacts with the nucleosome

Arp6 and Swc6 form a tight heterodimer withSwc6 entwined around the actin fold of Arp6that forms extensive contacts with two adja-cent OB-folds of the RuvBL hexamer, with den-sity for an ADP•BeF3 moiety at its center (fig.S3). This module provides a further tether tothe nucleosome via contacts made with nucleo-

somal H2A and linker DNA on the opposite sideto the Swr1 motor (Fig. 3, A to D). At this site,the DNA is unwrapped by ~65° from the cano-nical nucleosome path, exposing 2.5 turns of thewrap (Fig. 3B). This open conformation is stabi-lized via specific contacts with Swc6 (Fig. 3C)and may be further stabilized by Swc3 (fig. S6).In addition, a two-pronged interface of Swc6with H2A, involving a hydrophobic core centeredaround the aC region (Fig. 3D), could contributeto the specificity of SWR1 for H2A-containing

nucleosomes (30). Although the resolution ofthe interface between Swc6 and the nucleosomeis insufficient to identify the Swc6 sequence, theregion of contact with H2A involves residues 86to 94 and from residue 109 to the C terminus—regions with differences between H2A and Htz1that may contribute to the specificity. Interac-tions with Asp91 and Glu93 of the H2A acidicpatch anchor this module firmly on the face ofthe octamer. The acidic patch has previouslybeen shown to be important for SWR1 histone

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 3 of 8

Fig. 2. Details of theSwr1 motor andnucleosomal DNAdistortions at SHL2.(A) Top-down viewonto the nucleosomeshowing the positionof the Swr1 motordomain and theArp6/Swc6 module.Inset: Swr1 motordomain bound to thetracking strand (gold).(B) Nucleosomalposition of Swr1motor domain atSHL2, highlightingapproximate delinea-tions of HD1 andHD2 motor domainlobes. (C) Details ofthe Swr1 HD1 interac-tions with nucleoso-mal DNA. HD1 wedgesbetween the topand bottom gyres.(D) Details of theSwr1 HD2-DNA inter-actions. The DNA isdistorted in theSWR1-nucleosomecomplex as it liftsonto the HD2 motordomain. (E) Compari-son of Swr1 andChd1 nucleosomalDNA trajectories,showing Swr1 motordomain helices a9 anda10 (yellow) pushingon DNA. Structuresare aligned on themotor domainsof Swr1 and Chd1.The SWR1 footprint(17) coincides well withthe Swr1 binding site.

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

exchange (17). Similarly, we observe a loss inhistone exchange activity when Arp6/Swc6 isdeleted (Fig. 3E and fig. S6). Contacts madewith Arp6/Swc6 are mirrored by Arp5/Ies6 inthe INO80 complex (15, 16), which have beenproposed to facilitate a ratchet-like mechanismfor nucleosome sliding by INO80.A region of extended density, which we assign

as Swc2, forms a large interface crossing themain body of the complex, interacting with twoconsecutive subunits of the RuvBL hexamerbefore traversing the motor domains, crossingthe DNA wrap and then making contacts withthe distal face of the nucleosome on the acidicpatch (fig. S6). A strikingly similar region of den-sity is observed in INO80 corresponding to theIes2 subunit (16). Although these subunits donot share recognizable sequence similarity, boththeir structures and locations share much incommon. Ies2 plays an important role in regu-lating ATPase and sliding activities (31–33).However, although there are similarities in theinteraction (such as interaction with the acidicpatch), the actual binding site is quite different,so it is unclear whether the role of this interac-tion is similar or different in the two complexes.

SWR1 causes unwrapping of the DNA atthe nucleosome entry and exit sites

By creating nucleosomes with two overhangs(113N25), we satisfied the requirement of theSWR1 complex for at least a 60-bp overhang onat least one side for optimal activity, although asubstrate with overhangs on both sides appearsto bind even better (34). The structure shows in-teractions with both DNA overhangs (Fig. 3 andfig. S7). The DNA wrap at one end, correspond-ing to the SHL6-SHL7 region, is peeled backfrom the histone surface where it interacts withArp6/Swc6 (Fig. 3). DNA unwrapping is seen inmany other remodeling systems (INO80, Chd1),so it may be a common way to reduce friction asthe DNA wrap slides around histones (15, 16, 19).At the other DNA overhang, we also see inter-actions with the SWR1 complex (figs. S6 and S7).However, this interaction is rather variable andseveral different conformational classes could bedistinguished in the data set, suggesting a highdegree of mobility in this region (figs. S1 and S7).

Single-molecule assay to monitorSWR1-nucleosome unwrapping

To characterize SWR1-induced nucleosome un-wrapping observed in the cryo-EM structure, wedeveloped a single-molecule Förster resonanceenergy transfer (smFRET) assay to monitor nu-cleosome conformation upon SWR1 binding(Fig. 4A) (35, 36). Surface-immobilized fluores-cently labeled nucleosomeswere imaged by FRETin the absence and presence of SWR1. Nucleo-somes alone exhibit three main peaks (Fig. 4B),corresponding to bilabeled (~0.7 FRET), proximal-only (~0.9 FRET), and distal-only (~0.5 FRET)labeling, based on their photobleaching pattern(fig. S8). Counting the fraction of donor-acceptorlabeled nucleosomes as a function of time (Fig. 4C)shows that SWR1 remains as active on surface-

immobilized nucleosomes as in bulk (21). In thepresence of the slowly hydrolyzable ATP analogATPgS, exchange was not observed (Fig. 4C).This single-molecule histone exchange assay en-abled us to quantify whether either of the H2A/H2B dimers (distal or proximal) was exchangedpreferentially (fig. S8). The fraction of proximal-only fluorescent nucleosomes (Dp) remained con-stant over time (Fig. 4D).

SWR1 unwrapping is dynamicand stepwise

The single-molecule FRET trajectories reveal alarge increase in conformational dynamics in thepresence of SWR1 and ATP. In the absence ofSWR1, almost all observed trajectories appearstatic within our time resolution (250ms; Fig. 4Eand fig. S8), in agreement with previous studies(37, 38). In the presence of SWR1 and ATP, thefraction of dynamic traces increases to almost half(Fig. 4E). Dynamic FRET trajectories (Fig. 5A andfig. S9) show rapid excursions to lower FRETstates, consistent with SWR1-induced unwrap-ping of nucleosomal DNA as observed in thestructure (Fig. 3 and fig. S7). In the absence of

ATP, the fraction of static traces is comparableto that of nucleosomes alone (Fig. 4E), indicat-ing that ATP binding or hydrolysis is requiredfor unwrapping.A zoom into these excursions (Fig. 5B) reveals

the presence of various unwrapped states withFRET ratios ranging from 0.5 to 0.8. To analyzethe observed transitions, and to identify the var-ious observed states (Fig. 5A and fig. S9), we usedhidden Markov modeling (38). To simplify theanalysis, we focused exclusively on bona fideproximal-only fluorescent nucleosomes, althoughdistal-only fluorescent nucleosome trajectoriesexhibited similar behavior (fig. S10). A transitiondensity plot (TDP; Fig. 5C) summarizes the ob-served FRET transitions in all trajectories ana-lyzed. SWR1 inducesmultiple, partially unwrappedstates with the most frequent transition beingbetween the fully wrapped state (W, ~0.9 FRET)and the second unwrapped state (U2, ~0.7 FRET).Additional transitions to ~0.8, ~0.6, and ~0.5FRET indicate that unwrapping is a stepwiseprocess involving multiple intermediates. How-ever, most observed transitions occur directlywithout stopping at each intermediate state,

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 4 of 8

Fig. 3. Contributions of nonmotor subunits to structural changes in the nucleosome. (A) Overviewof SWR1-nucleosome structure showing the location of the Arp6-Swc6 module relative to Swr1motor domains. (B) Close-up of SHL6 region bound by Arp6-Swc6. The canonical nucleosome(PDB ID 1AOI; gray) is superimposed onto a Swr1-bound nucleosome. The change in anglecorresponds to approximately a ~65° rotation away from the nucleosomal wrap, exposing ~2.5 turnsof the canonical wrap. (C) Swc6 interactions pinning DNA in the SHL6-SHL7 region. (D) Swc6interactions with nucleosomal H2A showing specific interactions with the C-terminal tail andacidic patch (red). (E) The effect of subunit deletions on histone exchange activity of SWR1.Amino acid abbreviations: A, Ala; D, Asp; E, Glu; H, His; I, Ile; K, Lys; L, Leu; N, Asn; P, Pro; R, Arg;S, Ser; Y, Tyr. WT, wild type.

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

likely indicating that stepwise transitions be-tween intermediate states are faster than our250-ms time resolution. The symmetric shapeof the TDP about the diagonal shows that thesetransitions are reversible. Lowering the ATPconcentration to 100 mM results in similar butless frequent excursions (fig. S11), indicatingthat SWR1 can unwrap the DNAmultiple timesper encounter.

SWR1 unwrapping requires ATP bindingbut not hydrolysisIn the presence of a slowly hydrolyzable ATPanalog (ATPgS) that does not support histone ex-change over the time scale of these experiments(Fig. 4C), the fraction of dynamic traces remainshigh (Fig. 4E). The corresponding smFRET tra-jectories (Fig. 5D and fig. S12) exhibit similarrapid excursions into lower FRET states, indicat-

ing that nucleosome unwrapping requires ATPbinding but not hydrolysis. These excursions(Fig. 5E) involve similar transitions into vari-ous unwrapped states; however, steps intomulti-ple intermediate states are now apparent evenat 250-ms time resolution, likely due to slowertransitions in the presence of ATPgS. Likewise,the resulting TDP (Fig. 5F) reveals additionaltransitions to intermediate states (Fig. 5, C and

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 5 of 8

Fig. 4. Single-molecule nucleosomedynamics induced by SWR1. (A) Schematicdiagram of the experimental setup:Bilabeled nucleosomes (AF555 on eachH2A, blue), bound to biotinylated (orange)and labeled (AF647, red) 257-bp DNA, aresurface-immobilized on neutravidin-coated(yellow) biotin-PEG slides. Nucleosomes com-prise canonical H2A/H2B (beige discs) andH3/H4 dimers (gray discs). SWR1-nucleosomeinteractions are monitored via FRET betweenthe donors and the acceptor. (B) FRET histo-gram of nucleosomes (in the absence ofSWR1) shows three major populations: bila-beled (~0.7 FRET), proximal-only (~0.9 FRET),and distal-only (~0.5 FRET) fluorescentnucleosomes (N = 103). (C) Single-moleculehistone exchange assay shows that SWR1is active on immobilized nucleosomes at levelscomparable to bulk experiments (21) in thepresence of ATP (red, observed rate constantkobs = 0.05 ± 0.01 min−1) but not ATPgS (blue). Histone exchangeremoves fluorescently labeled H2A; therefore, activity is determined asthe fraction of remaining fluorescent nucleosomes as a function of time(N = 290 at t = 0 min in ATP or ATPgS). (D) The fraction of proximal-only (Dp) fluorescent nucleosomes remains constant with time, indicating that

proximal and distal H2A/H2B dimers are exchanged without preference(N = 74, 79, 36, and 40). (E) Percentage of dynamic traces in the presence ofnucleosome alone (NCP, N = 55); nucleosome and SWR1 complex(SWR1, N = 67); nucleosome, SWR1, and ATP (N = 85); ATPgS (N = 121);or ADP•BeF3 (N = 102). Error bars in (C) to (E) denote SE.

Fig. 5. Analysis of unwrapped states ofnucleosomal DNA. (A) FRET time trajectory(gray) of a proximal-only (Dp) fluorescentnucleosome in the presence of SWR1 and ATP,with resulting hidden Markov model fit (HMM,black). (B) Zoom of gray box in (A) showingdirect and reversible transient excursions intomultiple mid-FRET unwrapped states ({Ui}).(C) Calculated transition density plot (TDP) ofunwrapping of nucleosomal DNA in the presenceof ATP (2000 transitions from 60 trajectories).TDP confirms that multiple partially unwrappedstates are accessed from the high-FRETwrappedstate (W). (D) FRET trajectory (gray) of aproximal-only fluorescent nucleosome in thepresence of SWR1 complex and ATPgS withresulting HMM fit (black). Fewer and slowertransitions are observed than in (A). (E) Zoom ofthe gray box in (D) showing direct and reversibletransient excursions into multiple mid-FRETunwrapped states ({Ui}). (F) Calculated TDP ofunwrapping nucleosomal DNA in presence ofATPgS (1404 transitions from 65 trajectories).(G) Schematic of various nucleosomal unwrap-ping states.

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

G). Taken together, these data show that ATPbinding (but not hydrolysis) is required forSWR1 unwrapping on the pathway to H2A/H2B histone exchange. Similar results areobserved with ADP•BeF3 (Fig. 4E and fig. S13).To correlate the observed FRET values to the

number of unwrapped base pairs, we determinedan empirical distance calibration linking the dis-tance between the end of the DNA and the la-beled H2A histones (fig. S14). On the basis of thiscalibration, whichwe regard as qualitative rather

than quantitative, we estimate that SWR1 un-wraps 10 to 12 bp from the nucleosome, which isconsistent with the degree of unwrapping ob-served in the structure (Fig. 3 and fig. S7).

Exchange of histone dimers isdistributive with no preference forlinker-proximal or -distal dimers

Although dimer exchange by SWR1 is sequential(30), it is not known whether this process is dis-tributive or processive, nor is it known whether

there is any preference for which dimer is ex-changed first. We looked at histone dimer ex-change under catalytic conditions (1:10 ratio ofSWR1 to nucleosomes, i.e., 20 exchange reac-tions) at various ratios of dimer to nucleosome(fig. S15). At a 1:1 ratio, we observed around 31%doubly exchanged nucleosomes, although themajority (69%) have a single dimer exchanged,indicating a distributive rather than processivemechanism. At lower dimer/nucleosome ratios,the proportion of single exchanges is even higher(fig. S15). Furthermore, the proportion of dye-proximal and -distal dimer exchanges was ap-proximately equal, suggesting little preferencefor which dimer is exchanged first. Together,these data imply that SWR1 exchanges dimersone at a time, most likely binding in two orien-tations (probably at each SHL2 site) to achievethis process. The exchange reaction, therefore,is better considered in terms of which dimer isproximal or distal to the SWR1 motor domainbinding site rather than whether it is linker-distal or -proximal, because in vivo the nucleo-somes will be flanked by linker DNA on bothsides. As discussed above, the structure stronglysuggests that the dimer that is exchanged is theone proximal to the SHL2 site at which SWR1is bound, independent of which side any linkerDNA might be attached.

Implications for histone exchange

The mechanism for histone exchange by SWR1is still poorly understood. Very early studiesshowed that not only was the Swr1 subunit amember of the superfamily 2 translocases, likeother chromatin remodelers that slide nucleo-somes, but also that the ATPase activity of thissubunit was essential for activity (8); these find-ings suggested that DNA translocation at somelevel might be a component of the histone ex-changemechanism. However, no group has beenable to demonstrate net nucleosome sliding bySWR1, although limited sliding within the nu-cleosome has recently been reported (29).Experiments with nucleosomes containing

two-base gaps on the tracking strand withinthe SHL2 sites on either side of the nucleosomedyad have interesting effects on histone exchange(17). A two-base gap on one strand, at positionslocated between base pairs ~17 to 22 from thenucleosome dyad, permits exchange of one dimerbut prevents exchange of the other. Furthermore,a nucleosome in which gaps are placed at bothSHL2 regions (in both cases on the trackingstrand) prevents exchange of both histone dimers,even though binding to this nucleosome is un-affected. The region of SHL2 that is sensitive tothe presence of a gap (bases 17 to 22 from thedyad) corresponds very nicely with the contactregions of themotor domains with the DNAwrap(Fig. 2E). However, placing a gap on either side ofthis contact region, but on the same DNA strand,has only a modest effect on the rate of exchangeand both dimers can be exchanged. This latterobservation suggests that, at best, there can beonly limited translocation of DNA by themotordomains; similar experiments with bona fide

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 6 of 8

Fig. 6. Distortions of core histones as a consequence of DNA translocation. (A) Summary ofdistortions in the histone core. Histones are shown as cylinders, with coloring to distinguish betweencanonical and SWR1-bound nucleosome positions. Upper- and lower-tier histones were defined onthe basis of the views shown here. Canonical yeast nucleosome (PDB ID 1ID3) and SWR1-boundnucleosome were superimposed on residues 67 to 74 of H4, 108 to 126 of H2B, and 56 to 98 of H3.The general direction of histone movement is toward the motor-bound side of the nucleosome.(B) Edge-on view of the nucleosome, highlighting upper and lower histone tiers as well as upper andlower DNA gyres. Coloring is as in (A). Inset: Close-up of the motor-bound nucleosome region, showinga widening of the gap between the two gyres and distortion of the lower gyre in the presence ofSWR1. (C) Change in the upper-tier H2A-H3 interface. As a consequence of SWR1-nucleosome inter-actions, the upper H2A-H3 interface decreases by approximately 180 Å2, as determined in a com-parison with a canonical nucleosome (PDB ID 1ID3) (470 Å2 versus 650 Å2). By contrast, the lower-tierinterface remains largely the same (660 Å2 versus 650 Å2).These changes are further detailed in fig. S16.

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

nucleosome sliders show that a single-base gapahead of the motor domain binding site willblock translocation, but only when placed in thetracking strand (18). By contrast, for SWR1, a gapin the tracking strand close to the SHL2 site issufficient to prevent histone exchange, whichsuggests a requirement for some form of ATP-driven conformational changes of the DNA atthe binding site. Although footprinting showsthat there is no net sliding of nucleosome posi-tion as a consequence of exchange even at basepair resolution (17), limited transient transloca-tion (i.e., a few base pairs) of part of the DNAwrap has been proposed (29).In our structure, we do indeed observe limited

translocation of the DNA wrap from the entrysite to the Swr1 binding site by a single base pairto create a bulge. We also observe some distor-tion of the histone core (Fig. 6, fig. S16, andMovie 3). The histone core “flexes” and the uppertier, comprising a histone dimer and part of theH3/H4 tetramer, twists relative to the lower tier.These conformational changes result in changesat the interface between the H2A/H2B dimerand H3/H4 (Fig. 6 and fig. S16). The changes inthe nucleosome core likely also explain the in-creased dynamics of the DNA tails, as observedin our single-molecule studies (Figs. 4 and 5).Consequently, our structural and single-

molecule data suggest that the role of the Swr1ATPase activity is to induce local distortion inthe DNA at SHL2 that results from limited trans-location against the HD1 domain of Swr1, andthat this begins to destabilize the DNA wrap andhistone dimer contacts within the nucleosome.Such a distortion of the dimer interface with H4as a consequence of limited, and transient, DNAtranslocation has been suggested previously onthe basis of biochemical data (17, 29), and ourstructure now provides mechanistic insights intothat process. Our structure likely represents aninitial stage in the reaction cycle, and the dis-tortions we observe in both the DNA and thehistone core could be amplified by additionaltranslocation steps. Although ADP•BeF3 inducesconformational changes similar to those likelyaccompanied by ATP binding, we note that non-hydrolyzable ATP analogs do not support histoneexchange; thus, at least one, possibly several,rounds of ATP binding and hydrolysis may berequired. As a consequence of histone exchange,there need not be any net translocation of theDNA wrap beyond the Swr1 motor binding siteat SHL2, if any translocation of the DNA up tothat point were transient until histone dimerrelease and/or exchange.

Materials and methods

Briefly, nucleosomes and SWR1 complex wereprepared as described (21). The SWR1-nucleosomecomplex was assembled by incubating SWR1-ADP•BeF3:Htz1/H2B dimer complex with nu-cleosomes at a 1:1 molar ratio. Cryo-EM dataacquisition, image acquisition, and structure re-construction followed a similar procedure asdescribed (15). Data processing and refinementstatistics for the two cryo-EM structures are

summarized in figs. S1 to S3 and table S1. Seethe supplementary materials for further details,including details of single-molecule experiments.

REFERENCES AND NOTES

1. P. B. Talbert, S. Henikoff, Histone variants on the move:Substrates for chromatin dynamics. Nat. Rev. Mol. Cell Biol. 18,115–126 (2017). doi: 10.1038/nrm.2016.148; pmid: 27924075

2. M. Yun, J. Wu, J. L. Workman, B. Li, Readers of histonemodifications. Cell Res. 21, 564–578 (2011). doi: 10.1038/cr.2011.42; pmid: 21423274

3. S. J. Elsässer, S. D’Arcy, Towards a mechanism for histonechaperones. Biochim. Biophys. Acta 1819, 211–221 (2013).doi: 10.1016/j.bbagrm.2011.07.007; pmid: 24459723

4. B. Bartholomew, Regulating the chromatin landscape: Structuraland mechanistic perspectives. Annu. Rev. Biochem. 83,671–696 (2014). doi: 10.1146/annurev-biochem-051810-093157; pmid: 24606138

5. A. Jamai, R. M. Imoberdorf, M. Strubin, Continuous histoneH2B and transcription-dependent histone H3 exchange inyeast cells outside of replication. Mol. Cell 25, 345–355(2007). doi: 10.1016/j.molcel.2007.01.019; pmid: 17289583

6. O. Gursoy-Yuzugullu, N. House, B. D. Price, Patching BrokenDNA: Nucleosome Dynamics and the Repair of DNA Breaks.J. Mol. Biol. 428, 1846–1860 (2016). doi: 10.1016/j.jmb.2015.11.021; pmid: 26625977

7. C. R. Clapier, B. R. Cairns, The biology of chromatinremodeling complexes. Annu. Rev. Biochem. 78, 273–304(2009). doi: 10.1146/annurev.biochem.77.062706.153223;pmid: 19355820

8. G. Mizuguchi et al., ATP-driven exchange of histone H2AZvariant catalyzed by SWR1 chromatin remodeling complex.Science 303, 343–348 (2004). doi: 10.1126/science.1090701;pmid: 14645854

9. D. D. Ruhl et al., Purification of a human SRCAP complexthat remodels chromatin by incorporating the histonevariant H2A.Z into nucleosomes. Biochemistry 45,5671–5677 (2006). doi: 10.1021/bi060043d;pmid: 16634648

10. T. Kusch et al., Acetylation by Tip60 is required for selectivehistone variant exchange at DNA lesions. Science 306,2084–2087 (2004). doi: 10.1126/science.1103455;pmid: 15528408

11. M. Papamichos-Chronakis, S. Watanabe, O. J. Rando,C. L. Peterson, Global regulation of H2A.Z localization bythe INO80 chromatin-remodeling enzyme is essential forgenome integrity. Cell 144, 200–213 (2011). doi: 10.1016/j.cell.2010.12.021; pmid: 21241891

12. F. Wang, A. Ranjan, D. Wei, C. Wu, Comment on “A histoneacetylation switch regulates H2A.Z deposition by theSWR-C remodeling enzyme”. Science 353, 358 (2016).doi: 10.1126/science.aad5921; pmid: 27463665

13. S. Watanabe, C. L. Peterson, Response to Comment on“A histone acetylation switch regulates H2A.Z deposition bythe SWR-C remodeling enzyme”. Science 353, 358 (2016).doi: 10.1126/science.aad6398; pmid: 27463666

14. S. Brahma et al., INO80 exchanges H2A.Z for H2A bytranslocating on DNA proximal to histone dimers. Nat.Commun. 8, 15616 (2017). doi: 10.1038/ncomms15616;pmid: 28604691

15. R. Ayala et al., Structure and regulation of the humanINO80-nucleosome complex. Nature 556, 391–395 (2018).doi: 10.1038/s41586-018-0021-6; pmid: 29643506

16. S. Eustermann et al., Structural basis for ATP-dependentchromatin remodelling by the INO80 complex. Nature556, 386–390 (2018). doi: 10.1038/s41586-018-0029-y;pmid: 29643509

17. A. Ranjan et al., H2A histone-fold and DNA elements innucleosome activate SWR1-mediated H2A.Z replacement inbudding yeast. eLife 4, e06845 (2015). doi: 10.7554/eLife.06845; pmid: 26116819

18. A. Saha, J. Wittmeyer, B. R. Cairns, Chromatin remodelingthrough directional DNA translocation from an internalnucleosomal site. Nat. Struct. Mol. Biol. 12, 747–755 (2005).doi: 10.1038/nsmb973; pmid: 16086025

19. L. Farnung, S. M. Vos, C. Wigge, P. Cramer, Nucleosome-Chd1structure and implications for chromatin remodelling.Nature 550, 539–542 (2017). doi: 10.1038/nature24046;pmid: 29019976

20. X. Liu, M. Li, X. Xia, X. Li, Z. Chen, Mechanism of chromatinremodelling revealed by the Snf2-nucleosome structure.

Nature 544, 440–445 (2017). doi: 10.1038/nature22036;pmid: 28424519

21. C. L. Lin et al., Functional characterization and architecture ofrecombinant yeast SWR1 histone exchange complex. NucleicAcids Res. 45, 7249–7260 (2017). doi: 10.1093/nar/gkx414;pmid: 28499038

22. V. Q. Nguyen et al., Molecular architecture of theATP-dependent chromatin-remodeling complex SWR1.Cell 154, 1220–1231 (2013). doi: 10.1016/j.cell.2013.08.018;pmid: 24034246

23. R. J. Aramayo et al., Cryo-EM structures of the humanINO80 chromatin-remodeling complex. Nat. Struct. Mol. Biol.25, 37–44 (2018). doi: 10.1038/s41594-017-0003-7;pmid: 29323271

24. M. R. Singleton, M. S. Dillingham, D. B. Wigley, Structureand mechanism of helicases and nucleic acid translocases.Annu. Rev. Biochem. 76, 23–50 (2007). doi: 10.1146/annurev.biochem.76.052305.115300; pmid: 17506634

25. S. S. Velankar, P. Soultanas, M. S. Dillingham,H. S. Subramanya, D. B. Wigley, Crystal structures ofcomplexes of PcrA DNA helicase with a DNA substrateindicate an inchworm mechanism. Cell 97, 75–84 (1999).doi: 10.1016/S0092-8674(00)80716-3; pmid: 10199404

26. M. S. Dillingham, D. B. Wigley, M. R. Webb, Demonstration ofunidirectional single-stranded DNA translocation by PcrAhelicase: Measurement of step size and translocation speed.Biochemistry 39, 205–212 (2000). doi: 10.1021/bi992105o;pmid: 10625495

27. D. B. Wigley, G. D. Bowman, A glimpse into chromatinremodeling. Nat. Struct. Mol. Biol. 24, 498–500 (2017).doi: 10.1038/nsmb.3415; pmid: 28586327

28. O. Willhoft et al., Crosstalk within a functional INO80 complexdimer regulates nucleosome sliding. eLife 6, e25782 (2017).doi: 10.7554/eLife.25782; pmid: 28585918

29. R. K. Singh, S. Watanabe, O. Bilsel, C. L. Peterson, Transientkinetic analysis of SWR1C-catalyzed H2A.Z deposition unravelsthe impact of nucleosome dynamics and the asymmetry ofstepwise histone exchange. BioRxiv [preprint]. 19 April 2018.pmid: 304998

30. E. Luk et al., Stepwise histone replacement by SWR1requires dual activation with histone H2A.Z and canonicalnucleosome. Cell 143, 725–736 (2010). doi: 10.1016/j.cell.2010.10.019; pmid: 21111233

31. L. Chen, R. C. Conaway, J. W. Conaway, Multiple modesof regulation of the human Ino80 SNF2 ATPase by subunitsof the INO80 chromatin-remodeling complex. Proc. Natl. Acad.Sci. U.S.A. 110, 20497–20502 (2013). doi: 10.1073/pnas.1317092110; pmid: 24297934

32. W. Yao et al., Assembly of the Arp5 (Actin-related Protein)Subunit Involved in Distinct INO80 Chromatin RemodelingActivities. J. Biol. Chem. 290, 25700–25709 (2015).doi: 10.1074/jbc.M115.674887; pmid: 26306040

33. O. Willhoft, R. Bythell-Douglas, E. A. McCormack, D. B. Wigley,Synergy and antagonism in regulation of recombinanthuman INO80 chromatin remodeling complex.Nucleic Acids Res. 44, 8179–8188 (2016). doi: 10.1093/nar/gkw509; pmid: 27257055

34. A. Ranjan et al., Nucleosome-free region dominates histoneacetylation in targeting SWR1 to promoters for H2A.Zreplacement. Cell 154, 1232–1245 (2013). doi: 10.1093/nar/gkw509; pmid: 27257055

35. T. R. Blosser, J. G. Yang, M. D. Stone, G. J. Narlikar, X. Zhuang,Dynamics of nucleosome remodelling by individual ACFcomplexes. Nature 462, 1022–1027 (2009). doi: 10.1038/nature08627; pmid: 20033040

36. S. Deindl et al., ISWI remodelers slide nucleosomes withcoordinated multi-base-pair entry steps and single-base-pairexit steps. Cell 152, 442–452 (2013). doi: 10.1016/j.cell.2012.12.040; pmid: 23374341

37. G. Li, M. Levitus, C. Bustamante, J. Widom, Rapidspontaneous accessibility of nucleosomal DNA. Nat.Struct. Mol. Biol. 12, 46–53 (2005). doi: 10.1038/nsmb869;pmid: 15580276

38. S. A. McKinney, C. Joo, T. Ha, Analysis of single-moleculeFRET trajectories using hidden Markov modeling. Biophys. J.91, 1941–1951 (2006). doi: 10.1529/biophysj.106.082487;pmid: 16766620

ACKNOWLEDGMENTS

We thank Diamond for access and support of the Cryo-EMfacilities at the UK national electron bio-imaging center(eBIC), funded by the Wellcome Trust, MRC, and BBSRC.Funding: Supported by the Wellcome Trust [095519/Z/11/Z

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 7 of 8

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

and 209327/Z/17/Z (D.B.W.), Cancer Research UK (C6913/A21608 (D.B.W.)], the Medical Research Council [MR/N009258/1 and MR/R009023/1 (D.B.W.)], a core grant from the MRCLondon Institute of Medical Sciences (D.S.R.), and start-upfunds from Imperial College London (D.S.R.). Authorcontributions: O.W., M.G., D.S.R., and D.B.W. designed thestudies; O.W., E.Y.D.C., M.W., and R.A. performed the cryo-EManalysis; C.-L.L. and Y.C. conducted initial preliminary studies;O.W., E.A.M., and L.O. prepared the samples; M.G. conducted

and analyzed the single-molecule experiments; C.-L.L. andO.W. conducted the biochemical experiments; and D.B.W. andD.S.R. analyzed the data and wrote the manuscript with inputfrom all the authors. Competing interests: Authors declareno competing interests. Data and materials availability:Density maps are deposited at the Electron MicroscopyDatabase (accession code EMD-4395) and protein coordinatesare deposited at the Protein Data Bank (PDB ID codes 6GEJand 6GEN).

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/362/6411/eaat7716/suppl/DC1Materials and MethodsFigs. S1 to S16Table S1

References (39–48)

27 April 2018; accepted 8 August 201810.1126/science.aat7716

Willhoft et al., Science 362, eaat7716 (2018) 12 October 2018 8 of 8

RESEARCH | RESEARCH ARTICLEon M

arch 28, 2020

http://science.sciencemag.org/

Dow

nloaded from

Structure and dynamics of the yeast SWR1-nucleosome complex

Elizabeth A. McCormack, Lorraine Ocloo, David S. Rueda and Dale B. WigleyOliver Willhoft, Mohamed Ghoneim, Chia-Liang Lin, Eugene Y. D. Chua, Martin Wilkinson, Yuriy Chaban, Rafael Ayala,

DOI: 10.1126/science.aat7716 (6411), eaat7716.362Science 

, this issue p. eaat7716Sciencethe DNA was regulated by adenosine triphosphate (ATP) binding but did not require ATP hydrolysis.and the nucleosome destabilizes the DNA wrapped around the histone core. This SWR1-catalyzed partial unwrapping of

applied structural and single-molecule analyses to show that the interaction between SWR1et al.nucleosome. Willhoft containing dimer. Unlike all other remodelers, SWR1 does not translocate the−histone dimer for the Htz1 variant

The yeast SWR1 complex, a member of the INO80 family of nucleosome remodelers, exchanges the H2A-H2BFrom DNA unwrapping to histone exchange

ARTICLE TOOLS http://science.sciencemag.org/content/362/6411/eaat7716

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2018/10/10/362.6411.eaat7716.DC1

REFERENCES

http://science.sciencemag.org/content/362/6411/eaat7716#BIBLThis article cites 48 articles, 6 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on March 28, 2020

http://science.sciencem

ag.org/D

ownloaded from