Role of Residues 311/312 in Actin-Tropomyosin Interaction

According to the Lorenz et al. (Lorenz, M., Poole, K. J., Popp, D., Rosenbaum, G., and Holmes, K. C. (1995) J. Mol. Biol. 246, 108–119) atomic model of the actin-tropomyosin complex, actin residue Asp-311 (Glu-311 in yeast) is predicted to have a high binding energy contribution to actin-tropomyosin binding. Using the yeast actin mutant E311A/R312A in the in vitro motility assays, we have investigated the role of these residues in such interactions. Wild type (wt) yeast actin, like skeletal α-actin, is fully regulated when complexed with tropomyosin (Tm) and troponin (Tn). Structure-function comparisons of the wt and E311A/R312A actins show no significant differences between them, and the unregulated F-actins slide at similar speeds in thein vitro motility assay. However, in the presence of Tm and Tn, the mutation increases both the sliding speed and the number of moving filaments at high pCa values, shifting the speed-pCa curve nearly 0.5 pCa units to the left. Tm alone (no Tn) inhibits the motilities of both actins at low heavy meromyosin densities but potentiates only the motility of the mutant actin at high heavy meromyosin densities. Actin-Tm binding measurements indicate no significant difference between wt and E311A/R312A actin in Tm binding. These results implicate allosteric effects in the regulation of actomyosin function by tropomyosin.

The contraction of vertebrate striated muscle is regulated by the thin filament-associated proteins tropomyosin (Tm) 1 and troponin (Tn), which modulate the interaction of actin and myosin in a Ca 2ϩ -dependent fashion (2). Each Tm molecule associates with one Tn molecule and seven actin monomers. The amino acid sequence of Tm contains a pattern of charged and uncharged amino acids that repeats 14 times along its length (3). As each pair of repeats corresponds to an actin monomer along the actin filament, it has been inferred that the binding of Tm to actin is dominated by electrostatic interactions (1).
Tn is composed of three subunits: TnT, TnI, and TnC. The TnC subunit binds Ca 2ϩ and confers calcium sensitivity to the actin-Tm-Tn system. According to the three-state model of thin filament regulation (4), the actin-Tm-Tn complex can assume three structural states: blocked, closed, and open. In the blocked state there is a very low incidence of myosin binding. When Ca 2ϩ binds to TnC, the Tm-Tn complex shifts to the closed state, uncovering additional myosin weak binding sites on the actin filament. The increase in weak binding and the initial strong binding of myosin induce Tm-Tn to shift from the closed (where it prevents myosin strong binding) to the open state. According to this model, the azimuthal shift of Tm-Tn around the axis of the actin filament sterically regulates myosin binding to actin.
Despite the elegance of the three-state model, it cannot explain well the findings of earlier acto-S1 ATPase solution studies (5)(6)(7). Results from these studies describe the ability of Tm alone to both inhibit, at low S1 concentrations, actin-activated S1 ATPase rates and to potentiate the reaction at intermediate, nonsaturating concentrations of S1. Although inhibition can readily be explained using the steric block model, the potentiation suggests the presence of an allosteric component in actomyosin regulation.
The Ca 2ϩ -induced Tm-Tn movement on actin was first indicated by x-ray diffraction studies (8 -10). Electron microscopy has also been used to directly visualize this shift in Tm-Tn position (11)(12)(13)(14)(15)(16), and the studies of Limulus muscle (17,18) and vertebrate muscle (19) identified the positions for the Tm-Tn complex on actin in the presence and absence of Ca 2ϩ . A high resolution model of the Tm-F-actin complex was proposed by Lorenz et al. (1) on the basis of their x-ray fiber diffraction investigation. According to these authors, Tm alone and Tm-Tn in the presence of Ca 2ϩ reside in the same closedstate orientation on the actin filament. In this study Lorenz et al. (1) predict that 16 actin residues participate in the electrostatic interactions between F-actin and Tm and calculate their ⌬G contributions to this interaction. Our choice of residue 311 as a suitable starting point to test the predictions of this model and to gain additional insight into Tm regulatory function was based on two criteria. First, according to Lorenz et al. (1) residue 311 has a relatively high energy contribution to actin-Tm interaction, and secondly, there is a viable yeast actin mutant at this location: E311A/R312A (20) (Fig. 1).
In this study, we compared the regulation and function of wt and E311A/R312A yeast actins, as well as their interaction with Tm, in the in vitro motility assays, equilibrium binding experiments, and acto-S1 ATPase measurements. The results of our work support an allosteric explanation of the role of Tm in the regulation of actomyosin interaction.

MATERIALS AND METHODS
Reagents-ATP, ADP, dextrose, DTT, phalloidin, phenylmethylsulfonyl fluoride, and ␤-mercaptoethanol were purchased from Sigma. Yeast extract and tryptone were purchased from Difco. DNase I was purchased from Roche Molecular Biochemicals.
Proteins-Skeletal myosin and actin were prepared from rabbit back muscle according to Godfrey and Harrington (21) and Spudich and Watt (22), respectively. S1 and HMM were prepared from myosin using the protocol of Weeds and Pope (23) and Kron et al. (24), respectively. Yeast actins were purified over a DNase I affinity column (25) and were stored on ice in a G-actin buffer (5 mM TES, 0.2 mM CaCl 2 , 0.2 mM ATP, 0.5 mM ␤-mercaptoethanol, pH 7.6). Skeletal tropomyosin was prepared according to a previously reported protocol (26), as was cardiac troponin (27). N-(1-pyrenyl)iodoacetamide-labeled Tm was prepared using the protocol of Ishii and Lehrer (28) with a pyrene-to-Tm labeling ratio of 2.0. The cardiac troponin and tropomyosin were generous gifts from Dr. L. Tobacman. Yeast actin strains DBY6962 and DBY6958 (20), producing the E311A/R312A and K315A/E316A mutant actins, respectively, were generous gifts from Drs. D. Botstein and T. C. Doyle.
Actin-activated ATPase-The malachite green assay (29) was used to measure the ATPase activity of actin-activated S1. The procedure was the same as that used by Miller et al. (30). The assays were carried out at 22°C in a buffer containing 5 mM KCl, 2 mM MgCl 2 , and 10 mM imidazole, pH 7.4. The S1 concentration was 0.4 M, whereas that of the mutant or wt actin ranged between 0 and 35 M.
Regulated Actin-activated ATPase-Hydrolysis rates of regulated actin-activated S1 MgATPase at pCa values ranging between 5.0 and 9.0 were obtained by using light scattering to monitor the clearing time of regulated F-acto-S1 solutions. Clearing time is defined as the duration of the decrease in light scattering of acto-S1 solutions after the addition of ATP. The light scattering of the solution increases sharply upon the hydrolysis of the added ATP, leading to the determination of clearing time, i.e. the time of ATP hydrolysis. Thin filaments were reconstituted using either wt or E311A/R312A actin, bovine cardiac troponin, and either bovine cardiac or skeletal tropomyosin. The assay buffer was adjusted to 30 mM total ionic strength (including calcium concentrations) using a program written in QuickBasic by Drs. E. Homsher and N. Millar based on the equation of Fabiato and Fabiato (31). This buffer contained 5 mM imidazole (pH 7.5), 13.2 mM KCl, 3 mM MgCl 2 (free), 2 mM EGTA (with varying ratios of Ca 2ϩ -K ϩ -EGTA/K 2 EGTA), and 15 mM DTT. The S1 concentration was 1.0 M, whereas that of actin was 7.0 M. The concentrations of Tm and Tn were 2.0 and 1.2 M, respectively. Experiments were carried out at 23°C using a MgATP concentration of 0.1 mM. The course of MgATP hydrolysis was monitored by measuring the light scattering at 350 nm of the above solutions in a Spex Fluorolog (Spex Industries Inc., Edison, NJ).
Actin Polymerization-Polymerization of both the mutant and the wt G-actins (5.0 M) by MgCl 2 (3.0 mM) was monitored by measuring the light scattering in a Spex Fluorolog set at 325 nm.
Circular Dichroism and Tryptophan Fluorescence-The CD spectra of the monomeric actins were recorded between 190 and 250 nm in a G-actin buffer (see above) at 25°C using a Jasco J-600 CD spectropolarimeter. Tryptophan fluorescence emissions spectra of the G-actins were recorded at 23°C in Spex Fluorolog using 295 nm as the excitation wavelength.
Cosedimentation Assays-Using cosedimentation assays, we compared the strong binding of S1 (1-5 M) to each of the F-actins (4.0 M) stabilized by equimolar concentrations of phalloidin. The assays were performed at 22°C in a buffer made up of 3.0 mM MgCl 2 , 100 mM NaCl, and 10 mM imidazole at pH 7.4. The samples were pelleted using a Beckman TL-100 centrifuge and TLA-100 rotor spinning at 75,000 rpm (217,000 ϫ g) for 30 min. Prespun, supernatant, and resuspended pellet solutions were run on SDS-polyacrylamide gel electrophoresis and stained with Coomassie Blue. The densities of the various bands were determined using an Arcus II scanner (Agfa) and NIH Image version 1.60. The molar ratios of S1 bound to actin and the binding constants (K a ) were calculated by least squares regression using the program SigmaPlot 2.0 (Jandel Scientific). The binding of S1 to F-actin (3.0 M) in the presence of Tm (3.0 M) and MgADP (3.0 mM) was measured as above, except for the lower salt concentration used in the assay buffer (10 mM NaCl) because of the reduced S1 binding to actin in the presence of MgADP.
In addition, we compared the binding of Tm to the actins in the presence and absence of rigor-bound myosin S1. The assays were performed using 5.0 M actin (with equimolar concentrations of phalloidin) with and without 10 M S1, 0 -2.2 M N-(1-pyrenyl)iodoacetamidelabeled Tm in a 5.0 mM HEPES buffer, pH 7.5, 3.0 mM MgCl 2 , 150 mM NaCl at 22°C. The samples were pelleted as above. The concentrations of N-(1-pyrenyl)iodoacetamide-labeled Tm left in the supernatants were determined by fluorescence measurements using a Spex Fluorolog with excitation and emission wavelengths set at 344 and 385 nm, respectively.
In Vitro Motility Assays-The in vitro motility assays were performed according to a previously described protocol (30). The HMM titrations are a variation of the above protocol in that the HMM was applied to the nitrocellulose-treated coverslips at concentrations ranging between 0.06 and 0.3 mg/ml. Regulated thin filaments were assembled by incubating the rhodamine phalloidin-labeled actin filaments (2.0 M) with 0.5 M each of the regulatory proteins (Tm and Tn). The thin filaments were added to the motility assay coverslip at 10 nM. After a 1-min incubation, the unbound filaments were washed away with 50 l of an assay buffer containing 25 mM KCl, 1 mM EGTA, 4 mM MgCl 2 , 10 mM DTT, 0.1 M of both Tm and Tn (or Tm alone, depending on the nature of the assay), and 10 mM imidazole at pH 7.4. Regulatory proteins were included in the assay buffer to prevent the dissociation of these proteins from regulated actin (32). Movement was initiated by applying the same assay buffer containing 1.0 mM ATP and an oxygenscavenging system (33). An ExpertVision System (Motion Analysis, Santa Rosa, CA) was used to quantify the sliding speeds of individual filaments. Individual filaments were judged to be moving smoothly and were used for statistical analysis if the standard deviation of their sliding speeds was less than one-third of their average velocity (34).

Ca 2ϩ Titration of the Regulated System in the in Vitro
Motility Assays-According to Lorenz et al. (1), residue 311 has a considerable electrostatic role in the interaction between actin and Tm in the closed state (Ϫ850 cal/mol). If this is the case, replacing the charged glu-311 with an uncharged alanine could partially destabilize the closed state, causing changes in the regulation of the E311A/R312A actin mutant.
To test this possibility we measured the regulated thin filament sliding speed at various pCa values for both the regulated mutant and wt actins. Our results show a definite shift to the left of the pCa curve for the 311/312 mutant, indicating a lowered dependence on calcium to turn on the system ( Fig. 2A). We also found a similar trend in the numbers of moving filaments. At pCa 5.0 both actins have approximately the same fraction of motile actin filaments. At lower calcium concentrations, the regulated mutant actin has a larger percentage of moving filaments than does wt and ceases to move at a calcium concentration approximately half a pCa unit higher than that of wt actin (Fig. 2B). Experiments similar to those shown in Fig. 2 were repeated on four separate preparations of actins. In each case the pCa shift of the mutant versus wt complex was clearly defined. Although the midpoints of pCa titrations varied somewhat among the preparations, the shifts between mutant and wt actins were consistently reproduced (0.5 Ϯ 0.1 pCa units). Importantly, we saw no significant difference in sliding speed or percentage of filaments moving between the two actins in the absence of Tm and Tn (Table I).
As an additional control, we also performed motility experiments using the yeast actin mutant K315A/E316A. According to the calculations of Lorenz et al. (1), the electrostatic contribution of residue 315 is about 60% smaller than that of residue 311 (Ϫ330 versus Ϫ850 cal/mol). Thus, any changes in the regulation of this mutant actin should be smaller than those observed with E311A/R312A actin. Indeed, the pCa titration results of the 315/316 mutant (both the sliding speeds and percentage of filaments moving) fell between those of the other two actins (data not shown).
In Vitro Motility of Actin-Tropomyosin Complexes-Solution studies have shown that the presence of Tm strongly influences the acto-S1 ATPase rate; at low S1 concentrations Tm inhibits the ATPase reaction, whereas at high S1 concentrations the reaction is potentiated (5-7). According to biochemical and structural studies (1,4), the actin-Tm complex populates only two states, closed and open, of the three states proposed for the Tm-Tn-actin complex in the McKillop and Geeves (4) model. Clearly, the equilibrium between the closed and open states of actin-Tm is shifted by the binding of S1. When S1 is present at sufficiently high concentrations, actin-Tm is switched to the open state, releasing the inhibition of acto-S1 ATPase. To shed light on these transitions and on the change in Ca 2ϩ regulation of the regulated mutant actin, we measured the in vitro motilities of actin-Tm at various densities of HMM on the coverslips used in these assays.
In the case of wt yeast actin we observed a Tm-induced slowing of actin movement at low HMM concentrations (Fig. 3) but did not detect an acceleration of actin sliding by Tm at high HMM concentrations. Tm did not increase the sliding speed of wt thin filaments even when the switching "on" of the actin filaments was facilitated by increasing the HMM concentration to 0.5 mg/ml or by adding N-ethylmaleimide modified myosin S1 to the motility assay buffer. However, our results for mutant actin-Tm thin filaments (Fig. 3B) differ from those for wt actin (Fig. 3A). Binding of Tm to the mutant thin filaments significantly increased their sliding speed over surfaces incubated with HMM at concentrations greater than 0.12 mg/ml. At the highest HMM concentration tested, the mutant actin-Tm complex moved almost 50% faster than the mutant actin alone (5.7 versus 3.8 m/s). On the other hand, at low applied HMM concentrations (0.06 mg/ml), the mutant actin filaments behaved like their wt counterparts; the 311/312 actin alone was still sliding at speeds of approximately 1.8 m/s, whereas the actin-Tm complex did not move at all.
As expected, the percentage of filaments moving was high for both types of actin at high applied HMM concentrations (data not shown). On surfaces incubated with low HMM concentrations, however, the addition of Tm significantly reduced the numbers of moving filaments. This effect was especially marked for wt actin, with a significant drop in the numbers of moving filaments at 0.10 mg/ml HMM and no moving filaments noted at the lowest HMM concentration. The impact of Tm on the mutant actin was less pronounced, with a clear decrease in the number of moving filaments noted only at the lowest HMM concentration. Thus, it appears that lower concentrations of HMM are required to release the Tm-induced inhibition of actin sliding in the mutant than in wt actin filaments.
Binding of Tm to Actin and S1⅐ADP to Actin-Tm-In an effort to relate the observed effects of Tm on actin motility to the binding of Tm to actin, we measured the affinity of Tm for both the mutant and wt actins using cosedimentation assays. The binding of Tm to both actins was cooperative and could be described by sigmoidal curves with Hill coefficients of 2.4 Ϯ 0.4 for wt and 2.1 Ϯ 0.6 for 311/312 actin (Fig. 4). There was no significant difference in the binding coefficients, with K a values of (3.   (47), the strong binding constants of S1 to actin are about 10-fold lower for yeast than for rabbit skeletal actin.
the ⌬G of Tm binding to actin (1), it may actually impact this interaction. Thus, removing a positively charged residue (Arg-312) in addition to the negatively charged Glu-311 could decrease or obviate the effect of the 311 substitution on actin-Tm binding. To test the possibility that the mutant actin favors the binding of Tm in the open state, we also measured Tm binding to actin in the presence of rigor bound S1. Here, too, no significant differences between the 311/312 mutant and wt actins were noted under these conditions (data not shown). Finally, we checked whether the different effects of Tm on the sliding of wt and mutant actins in the in vitro motility assays might be related to different affinities of S1⅐ADP to their actin-Tm complexes. Binding experiments did not reveal any significant differences between the binding constants and Hill coefficients for S1-ADP interactions with wt and mutant actin-Tm (Table I).
Structure-Function Studies-Our results show significant differences in the motility of complexes of Tm-Tn (fully regulated) and Tm-alone with wt and 311/312 mutant yeast actin. Three possible explanations for these differences come to mind: (i) the mutations brought about changes in the actin structure which altered the function of actin, (ii) the affinity of Tn for Ca 2ϩ was impacted by these mutations in actin, or (iii) the structure was not disturbed, but the amino acid substitutions changed the interaction between actin and the regulatory proteins. We performed a series of experiments to test, and ultimately exclude, the first two possibilities.
Spectroscopic Assays-Structural properties of the wt and mutant actins were compared using CD and tryptophan fluorescence measurements. Far UV CD spectra of 311/312 mutant and wt actins are indistinguishable (Table I), indicating that the mutations have no discernible effect on the secondary structure of actin. Similarly, because the tryptophan emission spectra of the two actins are the same (Table I), the mutations do not appear to alter the tryptophan environments of the mutant actin.
Solution Interactions-The rate at which actin polymerizes from monomeric G-actin to polymeric F-actin is sensitive to its substructure and conformational state (35)(36)(37). We compared the rates of MgCl 2 -induced polymerization of both the 311/312 mutant and wt actins by measuring the increase in light scattering of 5.0 M G-actin solutions after the addition of 3.0 mM MgCl 2. We found no significant differences between the two actins ( Table I).
The interactions of the mutant and wt actins with myosin were compared in acto-S1 MgATPase measurements and in binding assays of the actins to S1 under rigor conditions. The acto-S1 Mg-ATPase activities for wt and 311/312 mutant actins were virtually identical, yielding almost the same K m and V max values (Table I). Thus, at the very least, the 311/312 mutations affect neither the weak binding interactions between actin and S1 nor the activation of S1 MgATPase by actin. Moreover, the in vitro motilities of the mutant and wt actins in the absence of regulatory proteins were the same, as was also the binding of S1⅐ADP to actin-Tm complexes containing either wt or mutant actin (Table I).
The strong binding interactions (i.e. in the absence of nucleotides) between each of the actins and S1 were examined using cosedimentation assays. Here too, the results did not reveal any significant difference between the two actins in their binding of S1 (Table I).
Finally, we compared the pCa dependence of regulated acto-S1 Mg-ATPase of both wt and 311/312 mutant actin to determine whether the mutation alters the calcium affinity of TnC in the regulatory complex (Fig. 5). We found no difference between the ATPases measured with wt or 311/312 mutant actin. Both had the same degree of activation versus pCa and the same maximal ATPase rates in experiments using bovine cardiac Tn and either bovine cardiac Tm (Fig. 5) or skeletal Tm (data not shown). We thus conclude that there is no difference in Ca 2ϩ affinity between thin filaments reconstituted with either of the actins.
All in all, the above assays suggest that there are no significant structural differences between wt and the 311/312 mutant actin. The results of the polymerization, acto-S1 MgATPase, regulated acto-S1 MgATPase, in vitro motility, S1 and S1⅐ADP binding, CD, and tryptophan fluorescence experiments strongly indicate that neither the structure nor the function of the mutant actin has been modified. Thus, the differences in the in vitro motility regulation that we report in this study probably stem from the differences in the interactions between the actins and the regulatory proteins.

DISCUSSION
The goal of this study was to test predictions of the Lorenz et al. (1) model of the actin-Tm complex in the closed state. According to this model, actin residue 311 contributes significantly to the closed-state interaction of actin and Tm. Our initial hypothesis, based on the above model, was that a mutation at this residue should reduce the affinity of Tm for actin by about 4-fold and thus destabilize the closed-state binding of Tm to the mutant actin. This, in turn, would be reflected in an altered Ca 2ϩ sensitivity of the regulated mutant actin filaments in the in vitro motility assays.
As a first step, we established that wt yeast actin is fully regulated by Tm-Tn (38). Thus, one important result of this work is that regulation can be conveniently studied using yeast actin in the in vitro motility assays. Korman and Tobacman (39) have also shown yeast actin to be fully regulated in acto-S1 ATPase studies. These results pave the way for regulation experiments using mutated yeast actins.
The main result of this work is that in the in vitro motility assay, the regulated 311/312 mutant actin filaments move faster at high pCa values than do regulated wt actin filaments. Because structural and functional comparisons of the two actins did not reveal any significant differences, these motility results indicate an increased Ca 2ϩ sensitivity of the regulated 311/312 actin filaments.
At first glance, these results appear consistent with the prediction of Lorenz et al. (1) regarding the role of actin residue 311 in actin-Tm binding. According to the McKillop and Geeves (4) three-state model, the binding of Ca 2ϩ to Tn induces conformational changes and steric transitions, which expose myosin weak binding sites on F-actin. At the same time the Tm-Tn complex moves into the closed position. If, in the mutant actin thin filaments, the Tm-Tn complex is not as firmly stabilized in the closed (ϩCa 2ϩ ) position, then a smaller number of myosin heads will be sufficient to induce a shift of the Tm-Tn complex into the open position. This should be especially apparent at low Ca 2ϩ concentrations, where few cross-bridges bind to the thin filaments. Our findings seemed to indicate that this is the case. In addition, the fact that larger fractions of mutant than of wt filaments moved at high pCa values lent support to this premise.
The explanation that the mutant actin regulation results are due, at least to some extent, to the destabilization of the actin-Tm complex in the closed state is not indicated by direct binding measurements. Tm binds to wt and 311/312 mutant actins with similar affinities under both closed-state (Tm alone) and open-state (Tm in the presence of S1 at a saturating concentration) conditions. Similar binding of S1⅐ADP to wt and mutant actin-Tm complexes rules out another possible cause for unequal stability of these actin-Tm complexes in the closed state. The unchanged binding properties of the 311/312 mutant suggest that explanations other than changes in equilibrium binding must be considered to account for our observations. These possibilities include: The The increased Ca 2ϩ sensitivity of the regulated 311/312 mutant actin may be because of a destabilization of the actin-Tm-Tn complex in the blocked state (ϩTn, ϪCa 2ϩ ), shifting the equilibrium toward the closed state. Such an effect could be the result of a reduced blocked-state affinity for Tm or an increase in the affinity of Tn for Ca 2ϩ in the mutant actin-Tm-Tn complex. Using regulated actin-activated S1 MgATPase assays, we showed that the 311/312 mutation does not change the affinity of Tn for calcium. It may appear surprising that the pCa profiles for regulated wt and mutant actin filaments are the same in ATPase activity measurements but different in the in vitro motility assays. However, the rate-limiting steps in the two types of experiments are different. Unlike the ATPase reaction, filament sliding is an analog of unloaded muscle fiber shortening and is rate-limited by ADP release from actomyosin⅐ADP (40). Consequently, ATPase values are not necessarily predictive of filament sliding speeds in the motility assays (41).
Regarding changes in blocked state stability, although we cannot exclude this possibility, circumstantial evidence argues against it. The modulation of actin motility by Tm alone shows that 311/312 mutation-induced changes also occur in the absence of the blocked state. Moreover, because actin residues 311 and 312 are not within the Tm binding site in the blocked state, any change in such binding to 311/312 mutant actin would be allosteric in nature.
Interpreting state of Tm on actin (1) could very well be an average of these Tm positions. If so, this would affect the identification of amino acid residues involved in the actin-Tm interaction and the estimation of their ⌬G contributions. An example of this would be the recent study of the actin mutation E93K in the Drosophila flight muscle (43). This residue was not implicated in the Lorenz et al. (1) study but, nevertheless, strongly affects the function of Actin-Tm filaments in the in vitro motility assay.
Finally, the 311/312 mutation may modify the regulation via allosteric shifts in the regulated actin system. One possible scenario is that the binding of S1 to the actin-Tm-Tn complex in the presence of ATP and at low calcium concentrations is enhanced by this mutation. Results of motility experiments with the actin-Tm (no Tn) complexes are also consistent with allosteric explanations. Inhibition of actin sliding by Tm at low densities of HMM on the motility assay surface could be described in purely steric terms, i.e. an insufficient density of myosin heads binding to the closed-state actin-Tm complex to tilt the equilibrium toward the open state (active form) of this complex. (This argument may also explain the prior observations that acto-S1 MgATPase is inhibited by Tm at low S1 concentrations (5-7).) However, the potentiation of the 311/312 mutant actin-Tm sliding at higher concentrations of HMM together with the reported potentiation of acto-S1 MgATPase by Tm as S1 concentrations are increased (5-7) (albeit not at conditions close to V max ) cannot be explained using a stericblock model of regulation or stronger S1⅐ADP binding to the mutant actin-Tm complex. These findings point to an allosteric change in the mutant actin-Tm complex leading to kinetic and/or mechanical consequences in the actomyosin cross-bridge cycle. Such consequences could, for example, include an increase in the rate of the ADP release step and/or of the ATP binding step and the subsequently more rapid release of the actin-Tm complex.
It should be noted that a missense mutation in actin (R312H) has been associated with hereditary idiopathic dilated cardiomyopathy, a heart failure of unknown origin (44). Nearly onethird of all heart failures are associated with relaxation abnormalities that could stem from an increased affinity of TnC for calcium, impaired sequestration of calcium by the sarcoplasmic reticulum, or slowed extrusion of calcium by the Na ϩ /Ca 2ϩ exchanger (45,46). The elevated pCa 50 for the speed-pCa curve ( Fig. 2A) suggests that regulated thin filaments containing the E311A/R312A actin mutant would exhibit slowed or impaired relaxation following contraction. A similar effect occurring in thin filaments containing the R312H mutation could contribute to the etiology of idiopathic dilated cardiomyopathy.
In summary, using in vitro motility assays, we found that yeast actin is a convenient substitute for skeletal actin in the study of regulation. Our in vitro motility experiments uncovered notable differences in regulation between wt and 311/312 mutant actin. As indicated by regulated acto-S1 ATPase assays, we saw no differences in the affinity of Tn for Ca 2ϩ in thin filaments reconstituted with either actin. We detected no significant differences between the two actins in their binding of Tm, S1, and S1⅐ADP and small, if any differences in other nonregulated functions. Our results underscore the importance of allosteric factors in the regulation of actomyosin interactions.