Fig 4. Myotilin regulates binding of tropomyosin to F-actin and interacts with α-actinin-2.

(A) Model of tandem Ig domains of myotilin bound to F-actin superimposed with the cryo-EM structures of cardiac tropomyosin bound to F-actin in the blocked (magenta ribbons, PDB: 5NOG), closed (cyan ribbons, PDB: 5NOL), and open (green ribbons, PDB: 5NOJ) structural states [35]. Residues of actin found in cross-links with N-terminal and carboxyl-terminal regions flanking myotilin Ig domains are shown in orange (see Fig 3). (B) Effects of myotilin on tropomyosin:F-actin interaction. Myotilin (Trx-MYOT) at increasing concentrations was added to F-actin before (Trx-MYOT before tropomyosin) or after (tropomyosin before Trx-MYOT) incubation of F-actin with the fixed amount of tropomyosin. Mean values (± SEM) of 3 independent experiments are shown. For “Trx-MYOT before tropomyosin,” the mean binding of tropomyosin to F-actin at all concentrations of Trx-MYOT was significantly different from the means in the absence of Trx-MYOT. Significance was assessed using 1-way ANOVA with Holm–Sidak test, p < 0.01 for 0 vs. 1.2 μM Trx-MYOT and p < 0.001 for 0 vs. all other concentrations of Trx-MYOT. For “Tropomyosin before Trx-MYOT,” the mean binding of tropomyosin to F-actin at all concentrations of Trx-MYOT, except for 1.2 μM Trx-MYOT, was significantly different from the means in the absence of Trx-MYOT. Significance was assessed using 1-way ANOVA with Holm–Sidak test, p < 0.05 for 0 vs. 2.4 μM Trx-MYOT, p < 0.01 for 0 vs. 3.6 μM Trx-MYOT, and p < 0.001 for 0 vs. 4.8 μM Trx-MYOT. (C) Schematic presentation of the ACTN2-WT, its constitutively open mutant (ACTN2-NEECK), and mutant of Trx-MYOT (Trx-MYOT-NEECK), possessing mutations resembling those in the ACTN2-NEECK (see Fig 4E). Lightning bolt depicts position of mutations. (D) Effects of myotilin on α-actinin–F-actin interaction. ACTN2-NEECK was added to F-actin at increasing concentrations before (ACTN2-NEECK before tropomyosin) incubation of F-actin with the fixed amount of tropomyosin. Mean values (± SEM) of 3 independent experiments are shown. The mean binding of tropomyosin to F-actin at all concentrations of ACTN2-NEECK was significantly different from the means in its absence. Significance was assessed using 1-way ANOVA with Holm–Sidak test, p < 0.05 for 0 vs. 1.2 μM ACTN2-NEECK, p < 0.01 for 0 vs. 2.4 μM ACTN2-NEECK, and p < 0.001 for 0 vs. all other concentrations of ACTN2-NEECK. (E) Sequence alignment of proteins and their residues involved in binding to α-actinin EF34. Residues of the CaM-binding motif 1-4-5-8 are shown in bold and boxed in green. Residues mutated in ɑ-actinin-2 NEECK and myotilin NEECK are shown in red. (F) Binding of myotilin to α-actinin. ACTN2-WT, ACTN2-NEECK, for details see (C and E), and ACTN2-EF14 were subjected at increasing concentrations to pull-down assay with (bound) or without (control) Trx-MYOT. The uncropped gels can be found in S1 Data. (G and H) Results of ITC experiments quantifying the interaction between (G), Trx-MYOT, or (H) its mutant (Trx-MYOT-NEECK), for details, see (C and E) and ACTN2-EF14. n.d., not determined. Data points that were used to create graphs are reported in S2 Data. See also S4 Fig. The 3D model presented in this panel is available in the following link: https://skfb.ly/6YGQt. ACTN2-EF14, α-actinin-2 EF14; ACTN2-WT, wild-type α-actinin-2; Trx, thioredoxin; WT, wild-type.