The Conserved Lysine-265 Allosterically Modulates Nucleotide- and Actin-binding Site Coupling in Myosin-2

Abstract

Myosin motor proteins convert chemical energy into force and movement through their interactions with nucleotide and filamentous actin (F-actin). The evolutionarily conserved lysine-265 (K265) of the myosin-2 motor from Dictyostelium discoideum (Dd) is proposed to be a key residue in an allosteric communication pathway that mediates actin-nucleotide coupling. To better understand the role of K265, point mutations were introduced within the Dd myosin-2 M765-2R framework, replacing this lysine with alanine (K265A), glutamic acid (K265E) or glutamine (K265Q), and the functional and kinetic properties of the resulting myosin motors were assessed. The alanine and glutamic acid substitutions reduced actin-activated ATPase activity, slowed the in vitro sliding velocity and attenuated the inhibitory potential of the allosteric myosin inhibitor pentabromopseudilin (PBP). However, glutamine substitution did not substantially change these parameters. Structural modelling suggests that K265 interacts with D590 and Q633 to establish a pivotal allosteric branching point. Based on our results, we propose: (1) that the K265-D590 interaction functions to reduce myosins basal ATPase activity in the absence of F-actin, and (2) that the dynamic formation of the K265-Q633 salt bridge upon actin cleft closure regulates the activation of product release by actin filaments.

Figure 1

Sliding velocity of actin filaments on WT and mutant myosin-decorated surfaces as determined by in vitro motility assays. (a–c) The histograms and Gaussian fits show typical distributions and average sliding velocities of actin filaments on lawns of WT compared to K265A, K265E and K265Q myosin lawns. (d) The bar graph represents the mean values of 7–10 different experiments with one or two protein preparations; the error bars show the standard deviation. The differences in sliding velocity obtained for the mutants were significant as compared to WT (*** for p < 0.0005).

Figure 2

Basal ATPase activity and basal P_i-release of WT and mutant myosins. (a) The basal ATPase rate was determined in the absence of actin. Error bars give the standard deviation for 3–6 independent experiments. The differences obtained for K265E and K265Q were highly significant (** for p < 0.005). (b) The basal P_i-release rate k _-4 is the rate limiting step of the basal Dd myosin-2 ATPase cycle and was measured using a fluorescent P_i detection system. The obtained kinetic parameters are summarized in Table 1(a) and Table 2(b).

Figure 3

Interaction scheme for actin and nucleotide binding of myosin. A refers to actin, M to myosin, T to ATP, D to ADP and Pi to inorganic phosphate. The asterisk (*) indicates the increased fluorescence of W501 induced by ATP hydrolysis. For the equilibrium binding constants a notation is used that distinguishes between constants in the absence and presence of actin by using italics type (k _x, K _x) and bold (k _x, K _x), respectively.

Figure 4

Steady-state ATPase activity of WT and mutant constructs. Determination of the ATPase activity for the WT and the three mutants from 0 to 80 µM F-actin. Data are mean values of 3–6 independent measurements and error bars are the standard deviation. Constants derived from this experiment are summarized in Table 1.

Figure 5

Binding and release of F-actin. Interaction of F-actin with WT and mutant myosin was followed using pyrene-labelled F-actin. (a) The actin concentration-dependent increase in the observed rate constants for pyrene-actin binding could be fit by a linear function. The slope defines the actin binding rate k _+A, which is slightly decreased by mutations K265E and K265Q. Error bars give the standard deviation for three independent experiments. (b) Fluorescence transients observed after chasing pyrene-actin from the complex with myosin by excess unlabelled F-actin. Single exponential fits to the data define F-actin release rate constants, which are significantly increased for all three mutant myosins. All resulting kinetic parameters for F-actin binding and release are summarized in Table 2.

Figure 6

Determination of ATP binding and hydrolysis rates for WT and mutant myosin constructs. The binding and subsequent hydrolysis of ATP by Dd myosin-2 induces conformational changes according to Fig. 3, which can be detected by a change of myosin’s intrinsic tryptophan fluorescence. The rate of ATP hydrolysis (k ₊₃ + k ₋₃) was decreased with high significance (p < 0.005) for K265E (see Table 2).

Figure 7

ATP-induced dissociation of acto-myosin complexes for WT and mutant myosin constructs. The dissociation of the rigor acto-myosin complex by increasing concentrations of ATP was followed by the exponential decrease in the light scattering signal. The plot of the observed rate constants versus the ATP concentration was fit by a hyperbola. For clarity, only the fits for WT and K265A are shown (see Table 2).

Figure 8

Actin activation of P_i-release for WT myosin and K265E mutant. (a) Averaged fluorescence traces for P_i-release at 25 µM F-actin are shown for the WT and K265E. (b) The P_i-release was determined in the presence of F-actin in a range of 0 to 25 µM. Error bars report the standard deviation and rate constants are listed in Table 2.

Figure 9

ADP release from myosin. (a) Averaged fluorescence traces of mant-ADP release after addition of excess (1 mM) ATP to a complex of 1 µM myosin and 20 µM mant-ADP (for clarity, only WT and K265Q data are shown). (b) The bar graph shows the ADP release rates and statistical significance for all four constructs. Error bars report the standard deviation and rate constants are listed in Table 2.

Figure 10

ADP inhibition of ATP-induced dissociation of acto-myosin complexes for WT and K265E. (a) The inhibition of the ATP-induced dissociation of the acto-myosin complex was followed by the exponential decrease in the light scattering signal. The graph shows sample traces for the WT at 50 and 800 µM ADP. (b) The plot of the observed rate constants versus the ADP concentration was fit by a hyperbola to give k _-AD and K _AD. The error bars report the standard deviation for three independent experiments (see Table 2).

Figure 11

Pentabromopseudilin-mediated inhibition of steady-state ATPase activities for WT and mutant constructs. Inhibition profile of the actin-activated ATPase activity (20 µM F-actin) of WT and mutant myosins by 100 µM PBP. Numbers on top indicate the extent of inhibition (n-fold change) compared to the 5% DMSO control. Data are mean values of 3–5 independent measurements and error bars give the standard deviation. The differences between DMSO control and PBP were significant in all cases (p < 0.05).

Figure 12

Structural model for the molecular basis of lysine-265 mutation-mediated effects on the allosteric pathway. (a) Overview picture of the myosin motor domain depicting the location of K265 in the rigor-like structures of Dd myosin-2 (blue, pdb: 2AKA) and of G. gallus myosin-5a (red, pdb: 1OE9). Note the completely closed actin-binding cleft in the myosin-5a structure and the partially closed cleft in myosin-2, which is proposed to close completely upon binding to F-actin. (b) Close-up view of the actin-binding cleft near K265 (Dd myosin-2) and K246 (Gg myosin-5a). (c) Schematic representation of the allosteric branching point K265/D590/Q633. The K265-Q633 interaction is established in the rigor state with closed actin-binding cleft. Red numbers indicate interaction distances in Å. The numbers in parentheses are distances proposed for the position of Gln633 in Dd myosin-2 with completely closed actin binding cleft (based on myosin-5a, pdb: 1OE9). (d–f) Illustration of the effects of mutations K265A, K265Q and K265E. Note the complete absence of interactions for Ala265, the repulsion between Glu265 and Asp590 and the retained Gln633 interaction of Gln265. (g) Schematic representation of interactions in the PBP binding pocket of WT Dd myosin-2 (based on pdb: 2JHR). (h) Mutation K265E eliminates specific interactions and thereby weakens the interaction network of PBP.