Distinct sites in tropomyosin specify shared and isoform-specific regulation of myosins II and V

Abstract

Muscle contraction, cytokinesis, cellular movement, and intracellular transport depend on regulated actin-myosin interaction. Most actin filaments bind one or more isoform of tropomyosin, a coiled-coil protein that stabilizes the filaments and regulates interactions with other actin-binding proteins, including myosin. Isoform-specific allosteric regulation of muscle myosin II by actin-tropomyosin is well-established while that of processive myosins, such as myosin V, which transport organelles and macromolecules in the cell periphery, is less certain. Is the regulation by tropomyosin a universal mechanism, the consequence of the conserved periodic structures of tropomyosin, or is it the result of specialized interactions between particular isoforms of myosin and tropomyosin? Here, we show that striated muscle tropomyosin, Tpm1.1, inhibits fast skeletal muscle myosin II but not myosin Va. The non-muscle tropomyosin, Tpm3.1, in contrast, activates both myosins. To decipher the molecular basis of these opposing regulatory effects, we introduced mutations at conserved surface residues within the six periodic repeats (periods) of Tpm3.1, in positions homologous or analogous to those important for regulation of skeletal muscle myosin by Tpm1.1. We identified conserved residues in the internal periods of both tropomyosin isoforms that are important for the function of myosin Va and striated myosin II. Conserved residues in the internal and C-terminal periods that correspond to Tpm3.1-specific exons inhibit myosin Va but not myosin II function. These results suggest that tropomyosins may directly impact myosin function through both general and isoform-specific mechanisms that identify actin tracks for the recruitment and function of particular myosins.

Keywords: actin cytoskeleton; actomyosin regulation; coiled coil; intracellular transport; muscle contraction.

Figure 1

Tropomyosin gene structure and mutations. (A) The exons expressed in Tpm1.1 (striated muscle αTm) and Tpm3.1 (Tm5NM1) from TPM1 and TPM3 genes, respectively. Colored blocks represent alternatively-expressed exons- orange in Tpm1.1 and blue in Tpm3.1. (B) Rat Tpm1.1 and Tpm3.1 amino acid sequences highlighting the sites of the mutations. Residues mutated in Tpm1.1 are orange; Tpm3.1 are cyan. The sequences are divided into seven (Tpm1.1) and six (Tpm3.1) periods according to the Phillips analysis [Phillips, 1986]. Each period corresponds to one actin subunit in the actin filament. The residue and period numbers are based on the Tpm1.1 sequence. All mutations were to Ala, except for Leu to Ser. The initial Met in Tpm3.1 is removed post-translation.

Figure 2

Regulation of skeletal myosin II and myosin Va-HMM by Tpm1.1. and Tpm3.1 mutants. The bars are grey for bare actin, blue for Tpm mutations that regulate both myosins, orange for mutations that have myosin-specific effects and black for WT Tpm and mutations with little to no effect. (A, B) Filament speeds for actin-Tpm1.1 or Tpm3.1 with skeletal myosin II from ensemble motility assays. The data for Tpm1.1-P3–P6 are from [Barua et al., 2012]. Values are mean ± s.d., n = 3–16 (Table 1). Assay conditions: 25 mM imidazole, pH 7.6, 25 mM KCl, 4 mM MgCl₂, 7.6 mM MgATP, 50 mM DTT, 0.5% methyl cellulose, and an oxygen scavenger system; 27°C. (C, D) Run frequencies of myosin Va-HMM moving on actin-Tpm1.1 or Tpm3.1. Run frequencies, defined as the number of moving motors per actin length per µM motor per time, are normalized to WT Tpm. Values are mean ± s.d., n = 6–17 (Table 2). Assay conditions: 10 mM imidazole, pH 7.4, 150 mM KCl, 4 mM MgCl₂, 1 mM EGTA, 50 mM DTT, 1 mg/ml BSA, 0.1 mg/ml CaMΔall, 1 mM MgATP, an oxygen scavenger system, and an ATP regenerating system; 23°C. *p <0.1, **p <0.01, ***p <0.001 compared to WT Tpm (ANOVA with Fisher’s LSD test)

Figure 3

Rate of dissociation of mdADP from actin-myosin-S1-mdADP-Pi or actin-myosin-S1-mdADP in the presence and absence of Tpm3.1. (A) k_fast is the rate of dissociation of Pi (A-MII-mdADP-Pi → A-MII-mdADP + Pi → A-MII + mdADP) measured by the rate of the decrease in mdADP fluorescence after dissociation from myosin II-S1. (C) The rate of Pi dissociation (k_obs) from A-MVa-ADP-Pi was measured directly using fluorescent phosphate binding protein. (B, D) Dependence of k_obs, the rate of dissociation of mdADP (A-M-mdADP → A-M + mdADP) on actin or actin-Tpm3.1 concentration. The dependence of the rates upon actin or actin-Tpm3.1 concentration were fit to the equation k_obs = k_max/ (1 + K_act/[actin]) to obtain values for the rate constants of Pi or mdADP dissociation at saturating actin concentration (k_max) and the binding affinities for actin or actin-Tpm3.1 (K_act). These values were used to calculate the second-order rate constants (k_max/K_act) of M-ADP-Pi or M-ADP binding to actin prior to product dissociation. In panel D, we have not reported K_act values since no data was collected at low concentrations of actin/ actin-Tpm3.1. Experimental conditions: 5 mM MOPS, pH 7, 2 mM MgCl₂, 1 mM DTT, 20 mM KAc; 20°C.

Figure 4

Summary of the effects of Tpm1.1 and Tpm3.1 mutants on the regulation of skeletal myosin II (filament speeds in ensemble motility assays) and myosin Va (run frequency in single molecule processivity assays). The values are percent change relative to WT Tpm. The boxes are shaded blue for Tpm mutations that regulate both myosins, orange for the mutations that have myosin-specific effects and white for little to no effect.

Figure 5

Molecular model for actin-tropomyosin-myosin in the open state. The model was constructed by docking a 7 Å-resolution Tpm1.1 crystal structure (Protein Data bank (PDB) ID 1C1G)[Whitby and Phillips, 2000] in a 8 Å-resolution rigor actin-Tpm-myosin S1 complex determined by cryo-EM (PDB ID 4A7L)[Behrmann et al., 2012]. The Dictyostelium myosin 1E (myoE) S1 structure in the EM model was replaced by myosin Va-S1 (PDB ID 1OE9)[Coureux et al., 2003] (A, B) and striated myosin II-S1 (PDB ID 4PA0) [Winkelmann et al., 2015] (C, D) structures through alignment with the upper 50K domain of myoE. A modification was made in the axial position of Tpm by a structural alignment of the Tpm1.1 crystal structure with the Tpm model in the actin-Tpm-myosin S1 complex, to optimize electrostatic interactions of the mutated residues in P3 and P6 with charged residues on myosin Va and myosin II, which resulted in an axial movement of the Tpm while maintaining the azimuthal alignment of Tpm on actin (see Materials and Methods). Tropomyosin is in cyan, actin in green and the myosin structures show the N-terminal 25K domain (blue), the converter domain and lever arm helix (yellow), the lower 50K domain (red) and the upper 50K domain (orange). Mutated residues in P3: (A) E114, K118 and D121 of Tpm can interact with K353, E369 and R343 of myosin Va, respectively, and (C) E114, D121 and E122 of Tpm can interact with K397, K367 and K365 of myosin II, respectively. Mutated residues in P6: (B) K233 and E240 of Tpm can interact with E369 and R343 of myosin Va, respectively, and (D) E234 and E240 of Tpm can interact with K397 and K365 of myosin II, respectively.