The actin-myosin interfaceMichael Lorenz1 and Kenneth C. Holmes

Biophysics Group, Max-Planck Institute for Medical Research, Jahnstrasse 29, 69120 Heidelberg, Germany

Edited* by James A. Spudich, Stanford University School of Medicine, Stanford, CA, and approved April 30, 2010 (received for review March 19, 2010)

In order to understand the mechanism of muscle contraction at theatomic level, it is necessary to understand how myosin binds toactin in a reversible way. We have used a novel molecular dynamicstechnique constrained by an EM map of the actin-myosin complexat 13-Å resolution to obtain an atomic model of the strong-binding(rigor) actin-myosin interface. The constraining force resulting fromthe EM map during the molecular dynamics simulation was suffi-cient to convert the myosin head from the initial weak-bindingstate to the strong-binding (rigor) state. Our actin-myosin modelsuggests extensive contacts between actin and the myosin head(S1). S1 binds to two actin monomers. The contact surface betweenactin and S1 has increased dramatically compared with previousmodels. A number of loops in S1 and actin are involved in establish-ing the interface. Our model also suggests how the loop carryingthe critical Arg 405 Glu mutation in S1 found in a familial cardio-myopathy might be functionally involved.

cryoelectron microscopy ∣ myosin II ∣ myosin V

The binding of myosin to actin can be weak or strong. Theaffinity, which changes over 5 orders of magnitude, is con-

trolled by ATP binding to the myosin head at a positionremote from the actin binding site. ATP binding produces “weakbinding.” In the absence of ATP the binding is “strong.” Themechanism of this interaction and its control by ATP is centralto an understanding of muscle contraction (1). Thus we needto know the structure of the strong-binding state of myosin toactin in atomic detail. Attempts to crystallize the complex havebeen unsuccessful. However, incubating myosin heads (subfrag-ment 1, S1) with filamentous actin (f-actin) produces “decoratedactin” in which each myosin head binds to the actin filament inthe strong-binding or “rigor” configuration. Decorated actin maybe used as a model of the actin-myosin “strong-binding” mode.Cryoelectron microscopy and 3D reconstructions of actin deco-rated with myosin II have produced a density map at 13-Å reso-lution (2). Combining this with structural data from the head ofmyosin V, which can be crystallized in the strong-binding myosinconformation (3), gave a near-atomic view of the actin-myosinrigor interaction (4). The actin binding site comprises parts ofthe upper and lower 50-K domains that are separated by a cleftin the weak-binding form. The cleft closes on strong binding.However, the interface appears incomplete: Some additional lo-cal refolding of myosin and actin appears to be necessary. Wehave now used constrained molecular dynamics—molecular dy-namics flexible fitting (MDFF) (5) to fit atomic models intothe interface using the EM density map as a constraint. In MDFFthe gradient of the density is used as an extra force field. Itappears that, in addition to the elements of the upper and lower50-K domains recognized in the earlier study, part of the flexibleloop 2 between the upper and lower 50-K domains in S1 and theactin-DNase I binding loop in actin form essential parts of theinterface. Furthermore, loop 4 in S1 (sequence 364–380) locatedclose to the cardiomyopathy loop (sequence 400–418) and loop 3(sequence 567–577) move in toward the actin binding interface,suggesting contacts with two actin monomers. Similar moleculardynamics studies were undertaken previously without EMrestraints (6). However, compared with the previous study ourstudies indicate about twice the area of interaction.

ResultsThe First Run with Unconstrained Coordinates. As starting coordi-nates for f-actin we used PDB ID code 2ZWH derived fromX-ray fiber diffraction data (7). For myosin II we used the crystal-lographic coordinates 2MYS (8). MD trajectories are often morestable with crystallographically derived coordinates. Since thecoordinates of f-actin were the output from an MD process,we replaced each of the two main domains of the actin monomercoordinates with the best overlaid coordinates from 1J6Z foreach domain (9) taking over just the DNAse I binding loopand C terminus from 2ZWH. This ensures good secondary struc-ture for the starting coordinates. Moreover, the 2MYS coordi-nates needed to be modified by removing the added methylgroups from the surface lysines [2MYS crystal structure wassolved with methylated surface lysines (8)] and by adding themissing side chain coordinates to the light chain Cα coordinates(this has no effect on the actin-myosin interface). Initially, theatomic coordinates were fitted to the EM density as rigid bodiesby least squares. Subsequently an MD trajectory was run for 5 ns.The 2MYS coordinates of myosin II are in the postpower strokeweak-binding form, whereas the EM density data are derivedfrom the rigor-like strong-binding form (4). Therefore, we antici-pated quite big changes on running an MD simulation.

Indeed, under the influence of the strong-binding (rigor) EMdensity the myosin II coordinates deform to become very muchlike 1W8J of myosin V (10), which is already in the strong-bindingform. The actin binding cleft closes and the β-sheet (Fig. 1B)takes on the more twisted configuration shown by myosin V.In addition, part of loop 2 (see Fig. 1B) initially inserted asrandom coil was moved into the actin-myosin interface (residues633–640 and 643–646).

The final MDFF calculations were carried out in the absenceof water because embedding the acto-S1 complex in a water boxshowed the same overall features such as the closing of the actinbinding cleft. We found that calculations in vacuum were suffi-cient in order to uncover the main structure segments and loopsthat participate in the acto-S1 interaction. The map force scalingwas set to 0.2 in order to avoid it dominating the MD simulation.The final model used five actin monomers to allow the majorS1-binding actin monomer (AC3; see Fig. 1A for an explanation)all possible interactions with neighboring actins. The EM densityforce was not applied to the lever arm in S1 (residues 795–843),the regulatory light chain and the essential light chain. The resultsare shown in Fig. 2.

The first runs displayed a number of problems including atendency to shift the structure toward the actin helix axis causedby a small but significant radial fall off of the EM density. Alsothe limited resolution of the EM density map resulted in stronggradients near the edges of the molecule, which caused some

Author contributions: M.L. and K.C.H. designed research; M.L. performed research; M.L.and K.C.H. analyzed data; and M.L. and K.C.H. wrote the paper.

The authors declare no conflict of interest.

*This Direct Submission article had a prearranged editor.1To whom correspondence should be addressed. E-mail: Michael.Lorenz@mpimf-heidelberg.mpg.de.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1003604107/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1003604107 PNAS ∣ July 13, 2010 ∣ vol. 107 ∣ no. 28 ∣ 12529–12534

BIOPH

YSICSAND

COMPU

TATIONALBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0

unrealistic bending of structures. Moreover, parts of loop 2 werepushed into the actin-myosin interface rather indiscriminately.

Nevertheless, the final model we obtained showed a goodmatch with the myosin V motor domain structure (10). The over-lay shown (Fig. 3) was calculated in a highly extensible programfor interactive visualization and analysis of molecular structures,called Chimera (11), based on secondary structure elements,showing the converged myosin II structure in blue and the myosinV structure in red. The actin binding cleft in myosin II closed andthe β-sheet took on the more twisted conformation found inmyosin V that is thought to represent the strong-binding confor-mation of S1 (12).

The rmsd between the starting model and the converged mod-el was slightly more than 7 Å after 5 ns of MD simulation, usingbackbone atoms of five actin monomers and S1 for the RMSDcalculation (Fig. S1, SI Text). Even though the MDFF runsconverged after 1–2 ns, the refinements were extended to over5 ns to ensure convergence.

Modeling Myosin II in the Strong-Binding Form. To counter the pro-blems described in the previous section we first generated startingcoordinates of myosin II in the strong-binding form based asfar as possible on crystallographically derived coordinates. Thenharmonic constraints were applied to areas not close to the inter-face to prevent unwarranted movement.

Previous studies (12) showed that myosin V without boundnucleotide is in the strong-binding form. As described above,unconstrained refinement of myosin II starting in the weak-bind-ing state transforms into a myosin V strong-binding (rigor-like)state when exposed to the map force (see Fig. 3). Thus myosinV provides a good template for modeling myosin II in itsstrong-binding state. The head of myosin II can be divided intothree domains (upper 50-K, lower 50-K, and N-C domain) eachof which is very similar in structure to the corresponding domainin myosin V. The 50-K upper, 50-K lower, and N-C-terminaldomain of myosin II weak binding and myosin V strong bindingcan bematched separately onto one another with high coincidence(SI Text). It transpires that two rotational transformations of theindividual domains suffice to bring thewhole structure ofmyosin IIinto good coincidence with myosin V. By using these transforma-tions to generate starting coordinates we are able to preserve thecrystallographic quality of the data.

For myosin V we used chain A of 1W8J (10). The two mole-cules were brought to similar orientations by overlaying the lower50-K domains (Fig. S2, SI Text). Then the upper 50-K domain ofmyosin II weak binding may be fitted to myosin V strong binding

by a 22° rotation of the upper 50-K domain around an axis thatruns approximately along the β-sheet and is roughly at right an-gles to the helix axis.

The N-C-terminal domain may be treated in the same way.Superposition of the N-C-terminal domain from myosin II (yel-low, Fig. S3, SI Text) on myosin V (apple green, Fig. S3, SI Text) isachieved by a rotation of 8.1° around an axis that runs close tostrand 3 of the β-sheet. Because this axis lies close to the domainboundary, the rotation amounts to an 8.1° bending of the β-sheet.The converter domain carries the long light chain bearing leverarm of myosin. The orientation of the converter domain achievedby the process outlined above leads to a positioning of the leverarm that is in excellent agreement with the EM density map.

The two rotations described above produced a model ofmyosin 2 in the strong-binding form (Figs. S4 and S5, SI Text).

Adding the Missing Loops. Loop 1. Loop 1 in myosin II (199–219) isdisordered as in most myosin structures. The residues from 205–215 are missing in 2MYS (8) and thus were model built andinserted as a random coil and were not exposed to the map forceor harmonic constraints.

Loop 2. Loop 2, between the upper and lower 50-K domains, isdisordered in most myosin S1 structures. It is missing in2MYS from residues 627–646. However, initial studies describedabove showed that at least some of loop 2 is involved in the actin-myosin interface. The resolution of the map is too low to build themissing structure without initial modeling. We have constructedthe part of loop 2 that appears to be involved in actin binding byhand using the EM map and crystallographic data from dictyos-telium myosin II as a guide.

Loop 2 is of variable length and is rich in lysine and glycineresidues. At the C terminus of the loop (647 in skeletal myosinII; 635 in myosin V) is an invariant threonine at the N terminus ofhelix W (649–667). The eight residues on the N-terminal side ofthe threonine are conserved in myosin II. They are quite differentin myosin V (loop 2 is also much longer) and indeed in all othermyosins. The N terminus of loop 2 is marked by an invariant phe-nylalanine. Dictyostelium discoideum myosin II (Dd) is unusualbecause the loop 2 region is short—it has only 16 residues.The whole of loop 2 is visible in 2AKA (13) and shows a β-hairpin.The form taken may be indicative of a preferred fold of the poly-peptide chain. Therefore, we have built an extended polypeptidechain (working from C to N) toward the actin surface from thebeginning of the W helix into a type-2 β-bend. After the bend an

Fig. 1. (A) Starting structure of the actin-S1 complexwith five actin monomers (AC1–AC5) and S1. The col-ors in the S1 structure indicate the upper 50-K do-main (red), the lower 50-K domain (orange), theremaining parts including the converter and theSH3-domain and the lever arm (yellow) and thetwo light chains (brown) for the unconstrained MDruns as described in the text. Also shown is the EMdensity (gray surface). (B) Important segments andloops of S1, which are often referred to in the text.ELC: essential light chain, RLC: regulatory light chain.Generated with Chimera (11).

Fig. 2. (A) The starting structure of the acto-S1complex as described in the text, embedded in theEM map. (B) The converged acto-S1 complex after1-ns MD run in MDFF. It can be seen that the upper50-K domain closes. Both pictures show the mainS1-binding actin monomer as surface representationin dark green, and S1 is shown as a ribbon. Only themotor domain and the beginning of the lever arm areshown—color code as in Fig. 1B. The orientation islooking along the helix axis from the + end of deco-rated actin. Pictures generated with Chimera (11).

12530 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1003604107 Lorenz and Holmes

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0

extended polypeptide runs away from the actin surface towardthe invariant phenylalanine that marks the N terminus of loop2. This configuration lies entirely within that part of the EM den-sity that was empty in earlier fits. With the given build, the Phe646 that is invariant in myosin II interacts with the Pro 602 of the“strut” (14), which is common to all myosin II (see Fig. 4). Thismay be a stabilizing interaction. Furthermore, on strong bindingof skeletal myosin II to actin, the sequence between Gly 634 andGln 647 would become an ordered part of the actin binding site,in which case all the trypsin-sensitive lysine residues would be-come unavailable for tryptic cleavage. This would provide a niceexplanation of the protection of loop 2 from tryptic attack whenmyosin is bound to actin (15). Murphy and Spudich (16) alsofound that changes in loop 2 altered the affinity of S1 for actinboth in the absence and presence of nucleotide. Chaussepied (17)proposed that residues 633–642 in S1 are already essential for theweak-binding state of the acto-S1 complex.

Loop 3. Loop 3 was completed by inserting the three missing re-sidues 572–574. It is located in the lower 50-K domain of S1 fromresidues 567–577 and moves toward AC1 (first actin monomer;see Fig. 1A) during the MDFF refinement. This possible contactwas already suggested by an earlier EM reconstruction (18).

We refer to myosin II matched onto myosin V and with loopsrebuilt as described as “myo2-5.”

Refinement of Myo2-5. We used myo2-5 as a starting model forseveral MD runs in MDFF with slightly altered parameters. Inaddition to the secondary structure preserving energy functiondescribed in the Materials and Methods, we added harmonic con-straints to several areas of actin and S1. The harmonic constraintswere applied to the Cα atoms of the designated areas in themolecules and were slightly altered for the three different runsin MDFF. The scaling of the density force from the EM mapwas also changed. Parameters are shown in Table S1 (SI Text).

The MD runs in MDFF were carried out for 5 ns for models1 and 2 where the density force was set to 0.1 and for 3 ns formodel 3 where the density force was set to 0.2.

For all three models we calculated the acto-S1 contact surfaceand the potential intermolecular H bonds and potential electro-static interaction. The surface areas are shown in Table 1. S1not only makes major contacts with the actin monomer (AC3)but also with themonomer (AC1) below themajor actinmonomerin the actin helix (see Fig. 1A). The measurements of the contactsurface area are in good agreement with previous results (19). Thesolvent accessible surface areas were calculated with visual mole-cular dynamics (VMD) (20) using the “measure sasa” routine (20).The sasa-parameter “sradius,” which is the additional value inÅadded to the atomic radius in order to scan the solvent accessiblearea, was set to 1.4 Å.

Furthermore, to regularize the result produced by the uncon-strained refinement of the DNase I binding loop, the rathersimilar crystallographic coordinates of the loop from the actin:DNase I crystal structure 1ATN (21) were inserted into the actincoordinates. This enhanced the interface with S1 by moving theDNase I binding loop into the S1 contact area as is described inthe following section. All three models converged to a similarresult with the same loops and segments being involved in theactin-S1 interface.

The details of the residues involved in contacts between actinand S1 varied with the altered loop and harmonic constraint con-ditions, thereby providing an estimate of the accuracy of themethod. The rms deviations of the three models from each otherranged between 1.0 and 1.1 Å. From the initial MYS2 coordinatesof myosin II used in the unconstrained first runs, the values ran-ged between 4.7 and 4.9 Å (the rmsd values were determined byselecting the backbone atoms of significant secondary structureelements of the 50-K upper and 50-K lower domains, resultingin 165 residues). In the next section we describe the details ofmodel 1 where the nucleotide binding pocket of S1 was preservedby applying strong harmonic constraints to the P loop and switch1 (these already had the geometry of the strong-binding statebecause myosin II had been matched onto myosin V).

The rmsd of the actin structure from the initial starting coor-dinates was 1.3 Å, mainly resulting from the regions that makeactin-S1 contacts and actin-actin contacts. The rmsd of the heavychain of S1 from the initial starting coordinates was 2.6 Å.

The Actin-Myosin Interface of Model 1. The refinement of the acto-S1 complex using MDFF (5) shows strong interactions not onlyfrom S1 to AC3 (the major actin binding site) but also from S1 toAC1, which increases the contact surface and probably stabilizesthe strong-binding state. Fig. 5 shows the binding site of the acto-S1 complex in stereo.

Contacts of S1 with AC3. The binding of S1 to AC3 (see Fig. 1A)shows a stable contact surface with S1 of 1;385 Å2 as shown inTable 1. The contact surface indicates several potential electro-static interactions and H bonds. The interaction between thelower 50-K domain and actin described in ref. 4—helix and bulge

Fig. 3. Comparison of the converged structure of myosin II in blue and myo-sin V (chain A of 1W8J) in red shown as secondary structure cartoons. Theactin binding cleft in myosin II has closed during the refinement and theβ-sheet has become more twisted. Generated with Chimera (11).

Fig. 4. Loop 2 sequences (after ref. 25). Three selected myosin II sequencesand one myosin V are shown: Gg Sk—skeletal myosin II, Dd—Dictyosteliummyosin II, squid myosin II (26), and Gg m5—chicken brain myosin V. The endsof loop 2 are shown in bold type. The C terminus of loop 2 ends in helix W.The sequence numbers of the last listed amino acid are shown on the right.Note that loop 2 in myosin V is considerably longer than in myosin II. MyosinIIs have a conserved phenylalanine (red) not present in myosin V. Loop 2 isgenerally not visible in crystal structures of myosin S1. The exception is Dd,which is short and ordered in two crystal structures.

Table 1. Contact surfaces of the actin-myosin models

Myosin2-5 Model 1 Model 2 Model 3

Contact surfaceAC3-S1 (Å2)

945.0 1,385.0 1,429.0 1,295.0

Contact surfaceAC1-S1 (Å2)

242.0 595.0 587.0 358.0

Total contactsurface (Å2)

1,187.0 1,980.0 2,016.0 1,653.0

Surface areas are listed separately for interactions between S1 and themajor actin (AC3) and the actin located underneath the major actin (AC1).Surface is given in Å2.

Lorenz and Holmes PNAS ∣ July 13, 2010 ∣ vol. 107 ∣ no. 28 ∣ 12531

BIOPH

YSICSAND

COMPU

TATIONALBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0

528–544 (S1) and segments 349–353 and 146–149 in AC3–ismaintained. Potential candidates for H bonds between S1 andAC3 are mainly found in the loops of S1, namely, loop 2, loop4, and the cardiomyopathy loop (see Fig. 1B). Furthermore,the loop 541–545 (S1) shows possible H bonds to AC3.

Fig. 6A shows the major potential H bonds between S1 andAC3. In addition, two lysines in the modeled loop 2 show possibleinteractions from Lys 640 (S1) to Asp 25 (AC3) and from Lys 642(S1) to Glu 334 (AC3). Loop 4 running from 366 to 377 in S1 inthe form of a β-hairpin also shows a number of potential electro-static interactions with residues in AC3. Fig. 6B displays the con-tact area of AC3 and S1 with potential electrostatic interactions.

The cardiomyopathy loop makes a direct electrostatic interac-tion via Glu 411 (S1) to Lys 336 (AC3). Other hydrophobic inter-actions also occur. The critical Arg 405 Gln mutation (identicalwith the Arg 403 Gln mutation in human cardiac myosin), whichoccurs in a severe familial hypertrophic cardiomyopathy and re-sults in a decreased actin-activated myosin ATPase (22), is not inthe interface: It faces away from actin. However, it may make asalt bridge with Glu 631 in the now structured part of loop 2.Therefore, its affect on the actin-myosin association is probablyvia a stabilization of the folded conformation of loop 2. Liu andco-workers (6) found in a MD simulation of the acto-S1 structureof Holmes et al. (2) that Lys 415 in S1 forms a stable H bond withGlu 334 in AC3. In our refinement we did not find this salt bridge.However, the participating atoms are only 6 Å away from each

other. Instead we found a H bond from Asn 410 of S1 to Tyr337 in AC3. They also found possible contacts of loop 4 to actin(they called it “C-Loop”). Our results are in broad agreementwith the findings of the earlier MD simulation (6).

Furthermore, Onishi et al. (23) performed mutations of threeregions in heavy meromyosin (HMM) by phosphorylation of re-sidues 546–548 (hydrophobic region), residues 407 and 409 and412–414 in the cardiomyopathy loop, and finally residues 652 and653 in loop 2. They all have significantly lower actin-activated AT-Pase in HMM or even completely extinguish it. We found allthose mutations in the contact surface of acto-S1 with possiblecontacts to actin residues (see Table 2).

Contacts of S1 with AC1. S1 also makes significant contacts withAC1 (see Fig. 1A). Milligan (18) suggested that there might bean interaction between loop 3 (567–577 in S1) and actin. Thiswas also suggested by Mornet et al. (15) by zero-cross-linkingexperiments. Furthermore, Blanchoin et al. (24) also favoredthe binding of S1 to two actin monomers in their kinetic studies.TheEMmap (2) used in the present paper shows the same featuresas the Milligan map: a hole in the map and a “finger” below inter-acting with actin. Fig. 7 shows model 1 with five actin monomersand S1 in the color code as in Fig. 1B in a surface representationplus the EMmap. The hole in the surface representation matchesthe surface of the model rather well.We refer to this feature as the

Fig. 5. Stereo view of the contacts of the acto-S1 complex of model 1. AC1 is shown in dark green, AC3 in light green, and S1 in yellow. Loop 2 is highlighted indark blue, cardiomyopathy loop in red, loop 4 in pink, and loop 3 in orange. The DNase I binding loop in actin is highlighted in cyan. The orientation isapproximately at right angle with the actin filament axis. The labels show residue type, residue number, and chain identifier (A for actin, H for S1). Generatedwith Chimera (11).

Fig. 6. (A) Possible H bonds between S1 (yellow) and AC3 (green) with loop 4, the cardiomyopathy loop, and at the lower part of loop 2. The position of Arg405 is shown in cyan as van der Waals surface pointing away from actin (see text). (B) Electrostatic interactions between S1 (yellow) and the major S1 bindingactin monomer (AC3) in green. The main interacting partners in S1 come from loop 2, loop 4, and the cardiomyopathy loop (CM-Loop). The labels show residuetype, residue number, and segment (A for actin, H for S1). Generated with Chimera (11).

12532 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1003604107 Lorenz and Holmes

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0

“Milligan contact.” The model of Liu and collegues (6) could notaccount for the Milligan contact.

The initial MD trajectory moved the DNase I binding loopaway from the configuration given in 2ZWH to a configurationsimilar to that found in the crystal structure of the actin-DNase Icomplex 1ATN (21). Replacing the DNase I binding loop in actinwith the original loop from the crystal structure revealed aninteraction of the DNase I binding loop with S1. During the sub-sequent MD trajectory the crystal configuration was modified,but it retained the general shape of the DNase I binding loop.It showed strong interactions with S1 particularly through44–49 (AC1) with S1 543–554 (S1).

We also found candidate H bonds between AC1 and S1 in theS1 helix 543–554 to the DNase I binding loop in actin and fromloop 3 in S1 to residues Tyr 91 and Arg 95 in actin. Possible Hbonds are shown in Fig. 8A.

Fig. 8B shows the potential electrostatic interactions of AC1with S1. Lys 50 located in the DNase I binding loop in actin showsa weak interaction with Glu 576 from loop 3 in S1. However, themajor contacts are established by the Milligan contact from Glu576 (S1) to Arg 95 (AC1) and Lys 569 (S1) to Glu 99 (AC1).

Table 2 shows all potential electrostatic interactions and poten-tial H bonds between AC1 and S1 and AC3 and S1 during theMDFF refinement.

DiscussionMDFF (5) trajectories run on an actin-myosin complex using theEM map at 13-Å resolution (2) as a constraint showed the cleft-open weak-binding state (myosin 2 postrigor) changing into astructure very similar to the cleft-closed strong-binding state (myo-sin 5 rigor) (Fig. 3). Moreover, the results showed that part of the

disordered loop 2 was involved in the actin-myosin interface. TheMDFF method in the unconstrained run showed that the actinbinding cleft, which is open in myosin II, has to close into a myosinV-like structure in order to account for the strong-binding state.However, the procedure distorted some regions near the bound-aries of the map and a falloff of radial density in the EM mapproduced a force pushing all structures 1–2Å toward the helix axis.Therefore details of the unconstrained run are uncertain and arenot further discussed. In order to rectify these deficiencies, we tookas a starting structure a modified myosin 2 modeled on the crystal-lographic structure of myosin 5. In addition, loop 2 in S1 and theDNase-binding loop in actin (AC1)were rebuilt as described in thetext. For stability, harmonic constraints were applied to selectedareas not involved in the interface. Three MDFF trajectories withvarying starting parameters led to similar structures for the inter-face. The generated interface is extensive and stereochemicallyplausible.

We found that myosin S1 interacts with two adjacent actinmolecules. In addition to the established interactions betweenthe lower 50-K domain and actin described in ref. 4, most of loop2 is involved in establishing the association of S1 and AC3.Furthermore, the cardiomyopathy loop participates in the acto-S1 interaction.

We also found substantial contacts between AC1 and S1. TheDNase-binding loop showed strong contacts with S1 residues.Moreover, the MDFF refinement moved loop 3 toward the actinhelix axis resulting in substantial support of an additional sitepredicted by Milligan (18).

Our results are also in good agreement with the findings by Liuand colleagues (6). However, the contacts from loop 2 to actindiffer as the part of loop 2 that makes contacts with AC3 inour acto-S1 structure was modeled in a different way to be buriedin the actin-S1 contact surface. There is also a discrepancy of theburied solvent accessible surface area. Whereas in our structurethe total contact surface of AC1-S1 and AC3-S1 is 1980 Å2, Liuand colleagues measure only 463 Å2. But they also favor loop 2,loop 4, and the cardiomyopathy loop in S1 as important segmentsto establish contacts with actin. Their MD simulations could nottake account for the Milligan contact as shown in our acto-S1structure that was driven by the map force during the MDFFrefinement.

The resulting acto-S1 model was achieved by using harmonicconstraints in the regions of actin and S1, which do not contributeto the acto-S1 and also actin-actin interface. The reason was thatthe resolution is too low to let all of the structure move in themolecular dynamics process using MDFF. However, it seemsto be very likely that actin would not behave like a rigid bodyupon S1 binding, resulting in at least small movements of loopsand segments in order to enhance the interface. Our actin-S1

Fig. 7. Surface representation of the model 1 and the EM map at 13-Åresolution (orientation: at right angles to the helix axis). The cutoff levelof the map is chosen in order to show the hole in the density surroundedby regions of AC1 that bind to S1. Color code as in Fig. 1B, ELC in light brown,RLC in dark brown. Generated with Chimera (11).

Table 2. Possible electrostatic interactions and H bonds between S1 and AC3 and AC1 (see Fig. 1 for explanation)

Potential electrostatic interactions Potential H bondsAC3 – S1 AC1 – S1 AC3 – S1 AC1 – S1

ASP24 – GLU629 LYS50 – GLU576 GLY23 N – LYS637 O LYS50 NZ – ASN552 OD1ASP25 – LYS640 ARG95 – GLU576 ASP24 N – GLY635 O ARG95 NE – PRO570 OARG28 – GLU629 GLU99 – LYS569 LYS328 NZ – ARG371 O ARG95 NH1 – ALA575 OARG147 – GLU373 TYR337 OH – ASN410 OD1 ARG95 NH2 – LYS572 OGLU167 – LYS544 SER348 OG – LYS637 O GLY46 O – LYS553 NZASP311 – ARG371 THR351 N – PRO529 O GLY48 O – LYS553 NLYS328 – GLU372 THR351 OG1 – PRO529 O GLN49 OE1 – SER549 OGGLU334 – LYS642 GLY23 O – GLY638 N TYR91 O – ALA571 NARG335 – GLU372 ARG28 O – ASN410 ND2LYS336 – GLU411 SER145 O – GLY643 N

GLU167 O – LYS544 NZGLN314 OE1 – ARG371 NH1ILE329 O – ARG371 NH2GLU334 O – LYS642 NZ

Lorenz and Holmes PNAS ∣ July 13, 2010 ∣ vol. 107 ∣ no. 28 ∣ 12533

BIOPH

YSICSAND

COMPU

TATIONALBIOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0

model also shows some small movements of actin segments in theinterface region. Further details would require higher resolutionof the EM map.

Our results reveal additional acto-S1 interactions that provide aplausible model for the strong-binding state of S1 to actin. Sincethe EMmap has a resolution of only 13Å, which is too low to showatomic structure, the details of the present model need to berefuted or substantiated by additional experiments. Fortunately,the additional potential contacts suggest numerous biochemicalexperiments.

Materials and MethodsMDFF (5) is a MD simulation with each atom being subjected to an additionalforce derived from the gradient of the EM density. The EM density map fromHolmes (2) was used to provide the constraining force field. This procedureresults in atoms being moved into higher density areas of the EM map. Inaddition, we added another energy term to preserve secondary structureelements as an option in MDFF. Thus the total energy function we used is

UTOTAL ¼ UMD þ UEM þ USS;

where UMD is the MD energy function, UEM is the energy function corre-sponding to a force derived from the gradient of the EM density,

and USS the energy function used to preserve the secondary structureelements. A scaling factor adjusted by trial and error was applied locallyto UEM in order to avoid dominance of the force resulting from the EMdensity map.

Calculations were carried out both in vacuum and in a water box in whichthe protein complex had been embedded. The box was set up to have per-iodic boundary conditions. The MD runs were carried out until the rmsdbetween the refined structures and the initial coordinate set reached a stablevalue. We carried out 12 MD runs in vacuum and 3 MD runs in a water box.Convergence was generally reached after 1 to 2 ns of simulation dependingon the parameter settings. The following parameters were changed duringthe refinements:

• The map force scaling, varying between 0.1 and 0.5,• simulation using 2 actin monomers or 5 actin monomers,• different start configurations of loop 2,• strength of constraints used to preserve secondary structure elements,• harmonic constraints—applied to selected areas.

ACKNOWLEDGMENTS.We thank Dr. Elizabeth Villa for her support to set up apreliminary version of MDFF and the fruitful discussions with her on themanuscript. We also thank Dr. Anne Houdusse for her valuable commentson the manuscript.

1. GeevesMA, Fedorov R, Manstein DJ (2005)Molecular mechanism of actomyosin-basedmotility. Cell Mol Life Sci 62:1462–1477.

2. Holmes KC, Angert I, Kull FJ, Jahn W, Schroder RR (2003) Electron cryo-microscopyshows how strong binding of myosin to actin releases nucleotide.Nature 425:423–427.

3. Coureux PD, et al. (2003) A structural state of the myosin V motor without boundnucleotide. Nature 425:419–423.

4. Holmes KC, Schroder RR, Sweeney HL, Houdusse A (2004) The structure of the rigorcomplex and its implications for the power stroke. Philos Trans R Soc London Ser B359:1819–1828.

5. Trabuco LG, Villa E, Mitra K, Frank J, Schulten K (2008) Flexible fitting of atomicstructures into electron microscopy maps using molecular dynamics. Structure16:673–683.

6. Liu YM, Scolari M, Im W, Woo HJ (2006) Protein-protein interactions in actin-myosinbinding and structural effects of R405Q mutation: A molecular dynamics study.Proteins 64:156–166.

7. Oda T, Iwasa M, Aihara T, Maeda Y, Narita A (2009) The nature of the globular- tofibrous-actin transition. Nature 457:441–445.

8. Rayment I, et al. (1993) Three-dimensional structure of myosin subfragment-1: Amolecular motor. Science 261:50–58.

9. Otterbein LR, Graceffa P, Dominguez R (2001) The crystal structure of uncomplexedactin in the ADP state. Science 293:708–711.

10. Coureux PD, Sweeney HL, Houdusse A (2004) Three myosin V structures delineateessential features of chemo-mechanical transduction. EMBO J 23:4527–4537.

11. Pettersen EF, et al. (2004) UCSF chimera—A visualization system for exploratoryresearch and analysis. J Comput Chem 25:1605–1612.

12. Sweeney HL, Houdusse A (2004) Themotormechanism ofmyosin V: Insights for musclecontraction. Philos Trans R Soc London Ser B 359:1829–1841.

13. Reubold TF, Eschenburg S, Becker A, Kull FJ, Manstein DJ (2003) A structural model foractin-induced nucleotide release in myosin. Nat Struct Biol 10:826–830.

14. Sasaki N, Ohkura R, Sutoh K (2000) Insertion or deletion of a single residue in the strutsequence of Dictyostelium myosin II abolishes strong binding to actin. J Biol Chem275:38705–38709.

15. Mornet D, Bertrand R, Pantel P, Audemard E, Kassab R (1981) Structure of theactin-myosin interface. Nature 292:301–306.

16. Murphy CT, Spudich JA (1999) The sequence of the myosin 50–20 K loop affectsMyosin’s affinity for actin throughout the actin-myosin ATPase cycle and its maximumATPase activity. Biochemistry 38:3785–3792.

17. Chaussepied P (1989) Interaction between stretch of residues 633-642 (actin binding-site) and nucleotide binding-site on sekeletal myosin subfragment-1 heavy-chain.Biochemistry 28:9123–9128.

18. Milligan RA (1996) Protein-protein interactions in the rigor actomyosin complex. ProcNatl Acad Sci USA 93:21–26.

19. Jones S, Thornton JM (1996) Principles of protein-protein interactions. Proc Natl AcadSci USA 93:13–20.

20. Humphrey W, Dalke A, Schulten K (1996) VMD: Visual molecular dynamics. J MolGraphics 14:33–38.

21. Kabsch W, Mannherz HG, Suck D, Pai EF, Holmes KC (1990) Atomic structure of theactin:DNase I complex. Nature 347:37–44.

22. Roopnarine O, Leinwand LA (1998) Functional analysis of myosin mutations that causefamilial hypertrophic cardiomyopathy. Biophys J 75:3023–3030.

23. Onishi H, Mikhailenko SV, Morales MF (2006) Toward understanding actin activationof myosin ATPase: The role of myosin surface loops. Proc Natl Acad Sci USA103:13897–13897.

24. Blanchoin L, Fievez S, Travers F, Carlier MF, Pantaloni D (1995) Kinetics of theinteraction of myosin subfragment-1 with G-actin. Effect of nucleotides and DNaseI.J Biol Chem 270:7125–7133.

25. Yengo CM, Sweeney HL (2004) Functional role of loop 2 in myosin V. Biochemistry43:2605–2612.

26. Yang Y, et al. (2007) Rigor-like structures from muscle myosins reveal key mechanicalelements in the transduction pathways of this allosteric motor. Structure 15:553–564.

Fig. 8. (A) Potential H-bond contacts between S1 (yellow) and AC1 (green). (B) Potential electrostatic contacts between AC1 (green) and S1 (yellow). The labelsshow residue type, residue number, and chain identifier (A for actin, H for S1). Generated with Chimera (11).

12534 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1003604107 Lorenz and Holmes

Dow

nloa

ded

by g

uest

on

Aug

ust 6

, 202

0