• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Apr 18, 2006; 103(16): 6136–6141.
Published online Apr 7, 2006. doi:  10.1073/pnas.0601595103
PMCID: PMC1434513

Toward understanding actin activation of myosin ATPase: The role of myosin surface loops


To understand the complicated interplay when a traveling myosin head reaches interaction distance with two actins in a filament we looked to three myosin loops that early on exert their influences from the “outside” of the myosin. On these we conduct, functionally test, and interpret strategically chosen mutations at sites thought from crystallography to be a patch for binding the “first” of the two actins. One loop bears a hydrophobic triplet of residues, one is the so-called “loop 2,” and the third is the “cardiomyopathy” loop. So far as we know, the myosin sites that first respond are the two lysine-rich loops that produce an ionic strength-dependent weak-binding complex with actin. Subsequently, the three loops of interest bind the first actin simultaneously, and all three assist in closing the cleft in the 50-kDa domain of the myosin, a closure that results in transition from weak to strong binding and precedes rapid Pi release and motility. Mutational analysis shows that each such loop contact is distinctive in the route by which it communicates with its specific target elsewhere in myosin. The strongest contact with actin, for example, is that of the triplet-bearing loop. On the other hand, that of loop 2 (dependent on drawing close two myosin lysines and two actin aspartates) is probably responsible for opening switch I and uncovering the γ-phosphate moiety of bound ATP. Taking into account these findings, we begin to arrange in order many molecular events in muscle function.

Keywords: site-directed mutation, hydrophobic triplet, loop 2, cardiomyopathy loop, pyrene–actin fluorescence

Conventionally, the difference between the (reasonable) in vitro rate at which myosin catalyzes ATPase in the presence of actin and the (very poor) rate that it does so in its absence expresses “actin activation.” The simple expression also highlights the in situ circumstance in which a traveling myosin head reaches interaction distance to the two actins to which it will bind. However, the simple expression obscures the complicated interplay between the proteins that next ensues and is the subject of this article.

Here we are mainly concerned with the one of the two actins that the myosin loops contact. The chemical kinetic scheme describing ATP hydrolysis by actomyosin (AM) is shown in Scheme 1 (2, 3), where A, M, and Pi are actin, myosin, and orthophosphate, respectively. Binding of ATP to AM dissociates the myosin from actin. Hydrolysis of ATP occurs principally in the dissociated state of myosin. After hydrolysis actin rebinds to myosin to form the weakly bound AM.ADP.Pi complex. This complex then isomerizes to the strongly bound state. This isomerization requires major conformational changes (including cleft closure of the 50-kDa domain) that result in Pi release and the power stroke and are greatly activated by actin. Earlier we found that binding of Lys-576** –Lys-578 of myosin to anionic residues of the second actin is involved in forming the weakly bound complex, whereas hydrophobic interactions of a triplet (Trp-546–Phe-547–Pro-548) of myosin with the “first” actin are contributors in the later, Pi-releasing process (5). However, information about first actin binding and activation of myosin ATPase is now to be provided.

Crystallographic and cryo-EM images of “rigor” AM (1, 68) suggest that the first actin binds to a myosin patch that is flanked on three sides by the loops: Leu-521 to His-566 in the lower 50-kDa subdomain (including a hydrophobic triplet of residues), Leu-634 to Phe-656 at the 50/20-kDa domain junction (loop 2), and Pro-405 to Lys-416 in the upper 50-kDa subdomain (cardiomyopathy loop). Presently we report on the enzymatic and motor functions of many heavy meromyosins (HMMs) obtained by specifically mutating sites in these loops (Table 1). In addition, by using quenching of the pyrene–actin fluorescence as a sign of a special structural change at the C terminus of actin (9), we report that mutating the triplet (Trp-546–Phe-547–Pro-548) reduces the quenching, whereas in WT or in mutations elsewhere the quenching upon complexing with pyrene–actin is unchanged. These experiments lead us to think that, although each of the loops exerts an influence that assists cleft closure, the influence originating at the triplet has a different routing from those originating elsewhere.

Table 1.
Enzymatic and motor functions of WT and mutant HMMs and their interactions with pyrene-labeled actin


I. Actin-Activated ATPase Activity of HMM Is Decreased by Mutations at Three Myosin Surface Loops.

To determine whether the maximum velocity (Vmax) or the apparent dissociation constant for actin (Kapp) is affected by mutations at the three loops, the steady-state ATPase activities of phosphorylated WT and phosphorylated mutant HMMs were measured as a function of [actin]. Previously we reported that each of various HMMs mutated at the hydrophobic triplet had a greatly reduced activity compared with that of WT HMM (5). Here we again measure actin-activated ATPase activities of three phosphorylated mutant (W546A, F547A, and P548A; see Supporting Text, which is published as supporting information on the PNAS web site) HMMs, but we reduce [KCl] to 10 mM to measure accurately kinetic parameters (Fig. 1A). Because F547A has essentially no actin-activated ATPase activity even at this [KCl], its Vmax and Kapp values cannot be measured. However, we can measure Vmax and Kapp values for W546A and P548A. Vmax is reduced to 1/15th of the WT value (1.78 s−1·head−1) for W546A and 1/40th for P548A, with no change in Kapp (Table 1). As earlier (5), we assume that the actin–HMM affinity in the weakly bound AM.ADP.Pi complex is related to Kapp and that the rate-limiting, Pi-leaving step can be described by Vmax of the actin-activated ATPase activity. These results suggest that the reduced level of actin activation is mainly caused by the slowed, rate-limiting, Pi-leaving step of the actin-activated ATPase reaction, not by the weakened affinity of these mutants for actin.

Fig. 1.
Actin-activated Mg-ATPase activities of various mutant HMMs as a function of [actin]. (A) Mutant HMMs W546A (○), F547A ([open triangle]), and P548A (□). (B) Mutant HMMs K652A (○), K653A ([open triangle]), and K652A/K653A (□). (C) Mutant ...

Joel et al. (10) found that a phosphorylated double mutant HMM, K652A/K653A, had essentially no actin-activated ATPase activity. To examine in the phosphorylated state whether either one or both of the lysine residues in loop 2 are responsible for actin activation, we measure actin-activated ATPase activities of two single mutants, K652A and K653A, as well as that of K652A/K653A (Fig. 1B). Vmax is reduced to 1/10th of the WT activity by the K652A mutation and to one-third by the K653A mutation (Table 1). A more drastic reduction results from losing two positive charges from loop 2, suggesting that positive charges at 652 and 653 are both important for full activation by actin. Kapp values for two single mutant HMMs (0.17 mM and 0.14 mM for K652A and K653A, respectively) are somewhat higher than the value for WT HMM (0.066 mM), indicating that loss of two positive charges from loop 2 decreases the affinity for actin in the weakly bound AM.ADP.Pi complex. Therefore, we conclude that ionic interactions between two lysines of loop 2 and two aspartates of actin differently function in the early, AM.ADP.Pi-forming stage and in the later, Pi-leaving stage.

Residues of the cardiomyopathy loop are in three categories as regards effects of their mutations on the actin-activated ATPase activity (Fig. 1C and Table 1). In phosphorylated I407A, D412A, and V414A Vmax is significantly decreased, but Kapp is not changed by these mutations. However, with phosphorylated V409A Vmax is not changed, but Kapp is decreased to one-fourth of the WT value. We therefore suggest that the first three residues are important in setting the actin-activated ATPase rate during the rate-limiting step, whereas the last residue engages in the actin–HMM affinity in the weakly bound AM.ADP.Pi state but not in the rate-limiting, Pi-leaving event. With phosphorylated V413A neither Vmax nor Kapp is changed by the mutation, indicating that this valine is unimportant for the myosin function studied.

II. Intrinsic ATPase Activity Is not Affected by These Mutations.

We examined effects of these mutations on the intrinsic ATPase activities to know whether structural changes induced by the mutations are global or not. As shown in Table 1, activities for mutants were not changed over the range from 0.8 to 2 of the WT value. We therefore conclude that these mutations do not create a structural disturbance around the ATPase catalytic center.

III. Motility of HMMs Differing at Three Surface Loops Differs from WT HMM.

In vitro motility assays were performed to examine whether the mutations at three surface loops affect the ability of HMM to move actin filaments (Table 1). Essentially all actin filaments moved at an average velocity of 0.91 μm/s in experiments performed with phosphorylated WT HMM. However, most of the actin filaments did not move in assays with phosphorylated F547A or with phosphorylated K652A/K653A, and velocities of the few filaments moved were very slow (<1/100th and 1/40th, respectively, of the velocity for WT HMM). Under the same conditions essentially all actin filaments move in experiments with other mutant HMMs, but with various average velocities. Among these mutant HMMs, decreases in the velocity of actin-filament movements correlate well with decreases in Vmax of actin-activated ATPase activity (Table 1). This result suggests that the abnormal motor function of these mutants observed in the assay probably results from a rate change in the Pi-leaving step that occurs after the first actin is bound.

IV. All Mutants at Three Surface Loops Can Form a Rigor Complex with Pyrene–Actin.

To examine whether mutant HMMs can form a rigor complex with actin, turbidity measurements were performed by incubating WT and mutant HMMs with pyrene–actin, then mixing with excess ATP. The decrease in turbidity upon addition of ATP depended on the amount of acto–HMM rigor complex present before the ATP addition. For WT and mutant HMMs, the amplitude of turbidity decrease was linearly dependent on [HMM] until a plateau is reached (Fig. 2). Amplitudes at the plateau for all mutant HMMs were practically identical to that obtained with WT HMM, indicating that the ability of HMM to form a rigor complex in the absence of ATP was not impaired by these mutations, although the affinity between HMM and actin cannot be measured by this method.

Fig. 2.
Binding of actin to WT and mutant HMMs and quenching of pyrene–actin fluorescence upon titration with WT and mutant HMMs. [diamond with plus] and ♦ represent WT HMM. (A) Mutant HMMs W546A (○ and •), F547A ([open triangle] and [filled triangle]), ...

V. Quenching of Pyrene–Actin Fluorescence by Rigor Binding of HMM Is Affected by Mutations at the Hydrophobic Triplet, but not by Mutations at Loop 2 or at the Cardiomyopathy Loop.

Conventionally, quenching of pyrene fluorescence when HMM binds to actin, whose Cys-374 is labeled with pyrene–iodoacetamide (9), is often used to detect formation of a rigor complex. To test whether mutant HMMs and actin form the rigor complex we also measured the fluorescence of pyrene–actin in the presence of various concentrations of WT and mutant HMMs (Fig. 2). For both WT and mutant HMMs the fluorescence is linearly decreased until [HMM] reaches one mole head per one mole of actin monomer (at higher concentrations it is independent of [HMM]). However (surprisingly), the saturated level of quenching of pyrene fluorescence was 74.4% of the initial for W546A, 17.1% for F547A, and 60.9% for P548A, compared with 78.6% for WT (Fig. 1A and Table 1). These results suggest that the ability of HMM to bind actin is not impaired by these mutations, but rigor complexes produced with these mutant HMMs did not have a normal rigor-like structure. Interestingly, the reduction in the extent of quenching was correlated well, although not linearly, with losses of actin-activated ATPase activity and motility seen with the same mutant HMMs. On the other hand, the maximum extents of quenching of pyrene–actin fluorescence by HMMs mutated at two other loops were identical to that obtained with WT HMM (Fig. 2 B and C and Table 1). To explain these observations, we assume that the strongest interaction in closing the cleft in the 50-kDa domain is the hydrophobic binding force between the triplet of myosin and the first actin and that this cleft closure is coupled with a conformational change in the C-terminal segment of actin. As a result, among the triplet mutants the reduced extents of quenching of pyrene–actin fluorescence are well correlated with losses of actin activation and in vitro motility.

VI. Both ON and OFF Rates of HMM Binding to Pyrene–Actin Are Affected by Mutations at Loop 2, the OFF Rate (but not the ON Rate) Is Affected by Mutations at the Hydrophobic Triplet, and Neither the ON Rate nor the OFF Rate Is Affected by Mutations at the Cardiomyopathy Loop.

The actin–HMM affinity in rigor is too large to permit direct measurement of molar ratios after equilibration, but we could get an approximate affinity by asymptotically measuring separately the ON and OFF rates and taking their ratio (using pyrene–actin and an HMM). For the ON rate we mixed a chosen HMM and pyrene–actin in a 1:2 molar ratio in a stopped-flow apparatus. We found that, in terms of measurable fluorescence, as a function of time, F(t) = F + (F0F)/(1 + kforward × [A]0 × t), where F(t), F0, and F are fluorescence intensities at time t, 0, and infinity, respectively, [A]0 is the concentration of pyrene–actin at time 0, and kforward is a second-order rate constant of the forward reaction. The best fitting theoretical curve superimposes well on the data for WT HMM (Fig. 5 Upper). Negligible deviation from the straight line of the difference plot (Fig. 5 Lower) implies that the binding event is a second-order reaction such as [actin] × [HMM], although an HMM molecule contains two actin binding sites. Observed values of kforward for WT HMM or for hydrophobic triplet or cardiomyopathy mutants are all almost the same; however, the values for K652A or for K653A are smaller by one-third, and for K652A/K653A it is smaller by one-sixth of the WT HMM value (Table 1).

To measure the reverse reaction, we arranged to displace pyrene–actin from its complex with an HMM (HMM:pyrene–actin ratio of 1:2) by treating it with an excess of pure actin. To be well fitted displacements from WT HMM and all mutants tested (an example is shown in Fig. 6) have to be described by an equation, F(t) = F + (F0F)/(1 + kreverse × [MA2]0 × t)†† , where F(t), F0, and F are the same as described above, [MA2]0 is the concentration of the HMM–pyrene actin complex at time 0, and kreverse is a second-order rate constant of the reverse reaction. The amplitude of F is close to that seen on dissociation by ATP (data not shown). None of these transients was affected by varying [actin] (4.5–9.0 μM), indicating that the deviation from a single-exponential curve is unlikely to be due to an [actin] deficiency. The observed value of kreverse × [MA2]0 is 27 times the value for WT HMM (0.00015 s−1) for W546A and P548A and 153 times the value for WT HMM for F547A, but for K652A, K653A, K652A/K653A, and D412A it is 2.7, 1.2, 6.0, and 1.7 times the value for WT HMM (Table 1).

Understanding technical limitations and some assumptions (e.g., pyrene-labeled actin is not greatly different from actin), we arrived at an estimate of the affinities of various HMMs for actin by recording the measured values of their ratios, kforward/(kreverse × [MA2]0). Such a procedure shows that mutations of the hydrophobic triplet remarkably, and comparatively, weaken the affinity. This observation is much in line with earlier, less direct inferences drawn in this work. So we conclude that the strongest interaction is that between the triplet of HMM and the first actin in closing the 50-kDa cleft and thus stabilizing the rigor complex. The value of kreverse for mutants at the C terminus of loop 2 is much smaller than that for those at the hydrophobic triplet. Nevertheless, the affinity is greatly decreased, because the kforward value is also decreased by mutations at loop 2. We speculate that this is why the actin-activated ATPase activity is greatly decreased, although these mutants with pyrene–actin show the same quenching of fluorescence as WT HMM does.


We have reported that mutations at various loop sites (space-filled balls in Fig. 3) reduce actin-activated ATPase (see I in Results) and in vitro motility (see III in Results). Structural observations (1, 6) show that in rigor, after the weak to strong transition, the hydrophobic triplet interfaces with Tyr-143, Ala-144, Ser-145, Gly-146, and Ile-345 of actin, that the two lysines interface with the two aspartates of actin, and that Ile-407, Asp-412, and Val-414 of the cardiomyopathy loop interface with Pro-333, Tyr-337, and Glu-334, respectively, of actin. We speculate that the movements that end with these interfaces close the cleft in the 50-kDa domain. The triplet loop and the cardiomyopathy loop lie at the distal ends of the lower and upper 50-kDa subdomains, respectively, so it is plausible that their bindings to actin assist closure. Lys-652 and Lys-653, at the C terminus of loop 2, are at the N-terminal end of the HW helix. Because the HW helix has several contacts with the lower 50-kDa subdomain, we can think that this helix moves together with the subdomain as the cleft closes. However, we cannot say that mutation of one or two residues abolishes cleft closure altogether, because all HMMs tested can form rigor complexes in the absence of ATP (see IV, V, and VI in Results). We suppose that these mutants have somewhat defective closure manifested in slowed weak to strong transition and Pi release.

Fig. 3.
Potential actin-binding amino acid residues on the myosin head and conjugate residues on actin. Crystal structures of the motor domain of Dictyostelium myosin with no nucleotide and monomeric rabbit actin are adapted from refs. 18 and 19, respectively. ...

Abnormal closure is shown in triplet-compromised mutants by their reduced extent of fluorescence quenching when complexed with pyrene–actin (see V in Results) and by instability of their rigor complexes (see VI in Results). We suppose that such quenching results from a local change in the lower 50-kDa subdomain rather than in a global change like cleft closure. A local change around actin Tyr-143, Ala-144, Ser-145, Gly-146, and Ile-345 upon binding of the triplet of HMM also occurs, and it is transmitted to Cys-374 of pyrene–actin by a C-terminal strand (Fig. 3, blue ribbon). Although details of the movement in the lower 50-kDa subdomain remain unknown, we guess that perturbation of the triplet remotely influences the active site by means of the lower 50-kDa subdomain, because the reduced extents of quenching are correlated well with losses in actin-activated ATPase and in vitro motilities (see I, III, and V in Results). The destination of the influence traveling from the triplet may be switch II in the active site (Fig. 4), because it moves with the lower 50-kDa subdomain and gates ATPase activity (12). As remarked above, on complexing the quenching of the actin-attached pyrene fluorescence indicates a conformational change at the C terminus of actin. However, because of the separation distance [>10 Å in a rigor model (6)], there is unlikely to be a direct interaction with the triplet, nor is it that the C terminus is essential for actin activation, because activation persists even after removing two terminal residues (13).

Fig. 4.
Transmission of the perturbation initiated at the actin-binding site to the nucleotide-binding pocket. (A) The motor domain of Dictyostelium myosin with MgADP.VO4. The crystal structure is adapted from ref. 20. The view in this figure, which ...

Loop 2 and cardiomyopathy mutants, on binding to pyrene–actin, suffer Vmax (see I in Results) and motility (see III in Results) reductions, but the fluorescence that they exhibit is no different from that of WT systems (see V in Results), indicating that the influences that they initiate travel to the nucleotide-binding pocket while avoiding pyrene-sensitive routings. The two C-terminal lysines in loop 2 are very far from the N-terminal 25-kDa domain, but they connect with it via a long α-helix (Fig. 3, HW helix, red ribbon). So we assume that displacements of these lysines can influence the location and orientation of the 25-kDa domain. We know that the same lysines interact with the first actin, both in the AM.ADP.Pi-forming stage and later in the Pi-releasing stage, because both Kapp and Vmax values are affected by their mutations (see I in Results). Together with our previous observation (5), we can say that when a traveling M.ADP.Pi myosin head reaches interaction distance to the two actins, myosin loop sites that first respond are two lysine-rich loops (a small loop including Lys-576–Lys-578 for the second actin and the aforesaid C terminus of loop 2 for the first actin), and the product is the weakly bound AM.ADP.Pi complex, which is ionic strength-dependent. Furthermore, we speculate that as the cleft closes these lysines at the N terminus of the HW helix are drawn more closely to the actin Asp-24–Asp-25 than before because of deformation of the lower 50-kDa subdomain. This movement greatly distorts the seven-stranded β-sheet by pulling on the third strand along the HW helix (arrows in Fig. 3). Because the first four and the last three strands belong to different domains (25-kDa domain, green; upper 50-kDa domain, cyan), the distortion twists the two domains. The bound nucleotide is on a side of the 25-kDa domain, switch I and Arg-247 are on the upper 50-kDa subdomain, and switch II together with Glu-470 are on the lower 50-kDa subdomain (Fig. 4B). As a result of the twist (between the 25-kDa domain and upper 50-kDa subdomain), a salt bridge, Asp-247–Glu-470, between switch I and switch II moves and uncovers the γ-phosphate moiety of the bound nucleotide. Before the twist, the salt bridge over the moiety assures a long lifetime to AM.ADP.Pi; after the twist, release of Pi is no longer hindered. So mutations at loop 2 or at the cardiomyopathy loop may still perturb the lower 50-kDa subdomain, but distortion of the β-sheet no longer occurs, in which case actin activation is abolished.

The advent of high-resolution crystallography in muscle research by Rayment and other pioneers (13, 6, 14) has enabled several “first principles” accounts of component parts in the overall phenomenon. For example, we know now that in the “power stroke” a linear force issuing from the enzyme pocket is transmitted by the relay helix, begins to be converted by the SH1 helix, and is then fully converted into a rotation, which finally drives the lever arm (15). Although still unconnected, there are now also ideas about how myosin catalyzes ATPase in the pocket (12). We see the present findings and surmise an as yet incomplete filling of a gap: how myosin anchors to actin and how actin then prepares myosin for its catalytic and thrusting function.

Materials and Methods

ATPase Assays.

The intrinsic ATPase activity of phosphorylated HMM was measured at 25°C in an assay medium containing 0.24 mg/ml phosphorylated HMM, 80 mM KCl, 2 mM MgCl2, 20 mM Tris-HCl (pH 7.5), 0.5 mM DTT, and 0.5 mM ATP.

The actin-activated ATPase activity was measured at 25°C as a function of the concentration of actin, [actin], in an assay medium containing 10 mM KCl, 2 mM MgCl2, 20 mM Tris-HCl (pH 7.5), 0.5 mM DTT, and 1 mM ATP with 4 μg/ml chicken gizzard myosin light chain kinase, 1 μg/ml bovine testis calmodulin, and 0.05 mM CaCl2. After terminating reactions with perchloric acid, the amount of released orthophosphate was measured colorimetrically by using the malachite green reagent (16).

In Vitro Motility Assay.

The motility assay was performed as in ref. 17. Briefly, an anti-c-myc mouse monoclonal antibody (Invitrogen) was adsorbed to a nitrocellulose-coated glass surface in a flow cell created by a nitrocellulose-coated coverslip and an uncoated coverslip. Phosphorylated HMM with a myc tag was bound to the nitrocellulose-coated glass surface by means of anti-c-myc antibodies. Rhodamine–phalloidin-labeled F-actin was infused into the flow cell, and the sliding movement of fluorescently labeled actin filaments in the presence of ATP was observed with an IX70 inverted microscope (Olympus) equipped with epifluorescence optics and a rhodamine filter set. Fluorescent images were detected with an electron-bombarded-CCD camera (Hamamatsu Photonics).

Fluorimetry and Stopped-Flow Experiments.

Fluorimetry was performed with an F-4500 fluorescence spectrophotometer (Hitachi, Tokyo). Stopped-flow experiments were performed with an SF61-DX2 stopped-flow spectrophotometer (Hi-Tech Scientific, Salisbury, U.K.) by using a 75 W Xe/Hg lamp and a monochromator for excitation wavelength selection. The formation of rigor complexes was detected by measuring the fluorescence decrease upon the addition of mutant HMMs to pyrene-labeled F-actin. For fluorimetry, pyrene fluorophores were excited at 346 nm, and the emitted light from fluorophores was monitored at 387 nm. For stopped-flow experiments, pyrene fluorophore was excited at 363.5 nm, and the emitted light was monitored after passing through a KV 389-nm cutoff filter (Schott, Mainz, Germany). Software provided by Hi-Tech Scientific was used for curve fitting of the data.

Turbidity Measurements.

Turbidity measurements were also performed with an F-4500 fluorescence spectrophotometer, but with exciting light at 500 nm and emitted light at 500 nm.

Supplementary Material

Supporting Information:


We are grateful to Prof. K. Mihashi for guiding us in the preparation of pyrene–actin and Profs. I. Rayment and T. Burghardt for counsel and for significant improvements in our manuscript. This work was supported by a Special Coordination Fund for Promoting Science and Technology from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (to H.O.) and by National Science Foundation Grant MCB 9603670 (to M.F.M.).


heavy meromyosin


Conflict of interest statement: No conflicts declared.

Rayment et al. proposed that a myosin head contacts two actin monomers (1). A lysine-rich surface loop issuing from the 50/20-kDa junction and two other surface loops constitute a major binding site for an actin monomer (termed the first actin). A small surface loop including Lys-576 and Lys-578 constitutes another binding site for the adjacent actin monomer (termed the “second” actin) in the same actin strand.

**Myosin exhibits many species differences, but the residues of interest here are highly conserved. The chicken gizzard myosin sequence numeration used is in ref. 4. Note that Arg-247, Ile-407, Val-409, Asp-412, Val-413, Val-414, Glu-470, Trp-546, Phe-547, Pro-548, Lys-652, and Lys-653 correspond to Dictyostelium discoideum Arg-238, Ile-398, Ala-400, Asp-403, Leu-404, Val-405, Glu-459, Val-534, Phe-535, Pro-536, Lys-622, and Lys-623. A mutant is represented by a one-letter expression of the original amino acid residue before its sequence number and that of the mutant residue after its number.

††Criddle et al. (11) reported that the time course of displacement by pure actin is describable by one or two exponentials depending on whether the material treated is subfragment 1 or HMM. However, no molecular picture of this interesting behavior was provided. To interpret various results of ours (given in the text), we suggest that an HMM molecule globally dissociates only when its two heads are simultaneously released from an actin filament. If so, we can say that HMM, of which one head still binds to the filament, is easier to rebind both heads to the filament, even when treated with an excess of pure actin. This suggestion seems to be reasonable, because such an HMM still anchors to the filament by means of the other head. In these simplified circumstances, displacing pyrene–actin with pure actin can be described by a second-order reaction v = kreverse × [MA2] × [MA2].


1. Rayment I., Holden H. M., Whittaker M., Yohn C. B., Lorenz M., Holmes K. C., Milligan R. A. Science. 1993;261:58–65. [PubMed]
2. Bagshaw C. R., Trentham D. R. Biochem. J. 1974;141:331–349. [PMC free article] [PubMed]
3. Bagshaw C. R., Eccleston J. F., Eckstein F., Goody R. S., Gutfreund H., Trentham D. R. Biochem. J. 1974;141:351–364. [PMC free article] [PubMed]
4. Yanagisawa M., Hamada Y., Katsuragawa Y., Imamura M., Mikawa T., Masaki T. J. Mol. Biol. 1987;198:143–157. [PubMed]
5. Kojima S., Konishi K., Katoh K., Fujiwara K., Martinez H. M., Morales M. F., Onishi H. Biochemistry. 2001;40:657–664. [PubMed]
6. Holmes K. C., Angert I., Kull F. J., Jahn W., Schröder R. R. Nature. 2003;425:423–427. [PubMed]
7. Holmes K. C., Popp D., Gebhard W., Kabsch W. Nature. 1990;347:44–49. [PubMed]
8. Rayment I., Rypniewski W. R., Schmidt-Bäse K., Smith R., Tomchick D. R., Benning M. M., Winkelmann D. A., Wesenberg G., Holden H. M. Science. 1993;261:50–58. [PubMed]
9. Kouyama T., Mihashi K. Eur. J. Biochem. 1981;114:33–38. [PubMed]
10. Joel P. B., Trybus K. M., Sweeney H. L. J. Biol. Chem. 2001;276:2998–3003. [PubMed]
11. Criddle A. H., Geeves M. A., Jeffries T. Biochem. J. 1985;232:343–349. [PMC free article] [PubMed]
12. Onishi H., Mochizuki N., Morales M. F. Biochemistry. 2004;43:3757–3763. [PubMed]
13. O'Donoghue S. I., Miki M., dos Remedios C. G. Arch. Biochem. Biophys. 1992;293:110–116. [PubMed]
14. Suzuki Y., Yasunaga T., Ohkura R., Wakabayashi T., Sutoh K. Nature. 1998;396:380–383. [PubMed]
15. Ohki T., Mikhailenko S. V., Morales M. F., Onishi H., Mochizuki N. Biochemistry. 2004;43:13707–13714. [PubMed]
16. Kodama T., Fukui K., Kometani K. J. Biochem. (Tokyo) 1986;99:1465–1472. [PubMed]
17. Kojima S., Fujiwara K., Onishi H. Biochemistry. 1999;38:11670–11676. [PubMed]
18. Reubold T. F., Eschenburg S., Becker A., Kull F. J., Manstein D. J. Nat. Struct. Biol. 2003;10:826–830. [PubMed]
19. Otterbein L. R., Graceffa P., Dominguez R. Science. 2001;293:708–711. [PubMed]
20. Smith C. A., Rayment I. Biochemistry. 1996;35:5404–5417. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...