Structural basis for tunable control of actin dynamics by myosin-15 in mechanosensory stereocilia

Figures

• 

  Fig. 1.. Cryo–electron microscopy structures of actomyosin-15 complexes. (A) Diagram of Mus musculus myosin-15 primary domain structure. (B) Cryo–electron microscopy (cryo-EM) maps of F-actin alone and indicated actomyosin-15 complexes. (C) Cryo-EM map and ribbon diagram of two actin and one myosin-15 subunits from the rigor wild-type actomyosin-15 reconstruction. Numbered boxes correspond to detail views in subsequent panels. (D) Cryo-EM map and stick models around actin’s ADP binding pocket. (E) Cryo-EM map and stick models at the actin-myosin loop 3 interface. The position of the jordan mutation (D1647) is highlighted in teal.

• 

  Fig. 2.. Structural mapping of deafness causing myosin-15 mutations. (A) Ribbon diagram of the myosin-15 motor domain with α-carbons at sites of deafness-causing mutations (spheres), colored according to the indicated mechanistic categories. (B) Detail view of mutations anticipated to disrupt folding. (C) Detail view of mutation anticipated to affect subdomain rearrangements. (D) Detail view of mutations in the switch I, switch II, and P-loop regions anticipated to affect ATPase activity.

• 

  Fig. 3.. Interactions at the rigor wild-type actomyosin-15 interface. (A) Atomic model of the interface between two actin subunits and rigor wild-type myosin-15. Actin subunits i, i + 2, and myosin-15 are colored in light blue, cornflower blue, and dark orange, respectively. The indicated myosin-15 actin interaction motifs are highlighted in magenta, and numbered boxes correspond to detail views in subsequent panels. (B) Hydrogen bonding interactions between myosin-15 residue D1647 and actin residues S352 and T353. (C) Hydrophobic interactions formed between the D-loop of actin subunit i + 2, subdomains 1 and 3 of actin subunit i, and the HLH motif of myosin-15. (D) Interface between the CM loop of myosin-15 and subdomain 3 of actin subunit i. (E) Contacts between loop 3 of myosin-15 and subdomain 1 of actin subunit i + 2.

• 

  Fig. 4.. Myosin-15 binding remodels F-actin while maintaining D-loop flexibility. (A) Ribbon diagram of two actin subunits in the presence and absence of myosin-15, superimposed on subunit i. Minor D-loop conformations are colored in blue (actin alone) and pink (myosin-15 bound). (B) Structural comparison at the interface formed by two longitudinally adjacent actin subunits and the myosin-15 HLH. Bare F-actin and rigor wild-type (WT) actomyosin-15 are colored in light blue and dark orange, respectively. Minor D-loop conformations are colored in blue (actin alone) and pink (myosin-15 bound) and are shown in transparent representation. (C) Segmented cryo-EM density of the bare F-actin D-loop rendered at 0.036 RMS, superimposed with stick models of the Out and In D-loop conformations colored in light blue and medium blue, respectively. (D) Segmented cryo-EM density of the rigor wild-type myosin-15–bound actin D-loop, superimposed with stick models of the Out and In D-loop conformations colored in dark orange and pink, respectively. (E) Backbone representation of the In conformation of the indicated models, superimposed on the segmented cryo-EM density the of myosin-15–bound F-actin D-loop.…

• 

  Fig. 5.. Binding by rigor jordan mutant myosin-15 locks the actin D-loop. (A) Atomic model of the interface between two actin subunits and rigor jordan myosin-15. Actin subunits i, i + 2, and myosin-15 are colored in light blue, cornflower blue, and medium purple, respectively. Numbered boxes correspond to detail views in subsequent panels. (B) Actomyosin interface at the site of p.D1647G mutation. (C) Structural comparison of bare F-actin (light blue), rigor wild-type (dark orange), and jordan mutant (medium purple) myosin-15–bound F-actin at the interface formed by two longitudinally adjacent actin subunits and the myosin HLH. (D) Segmented cryo-EM density and atomic model of the rigor jordan mutant actomyosin-15 D-loop. (E) Backbone diagram of the three In D-loop conformation models superimposed on segmented cryo-EM density of the rigor jordan mutant actomyosin-15 D-loop.

• 

  Fig. 6.. ADP-bound myosin-15 undergoes minimal rearrangements and locks the D-loop in the In conformation. (A) Cryo-EM maps of indicated actomyosin-15 complexes derived from reprocessing with symmetry expansion and recentering on myosin-15’s residue S1245. (B) Superposition of rigor and ADP wild-type actomyosin-15 atomic models. Actins from the rigor atomic model are shown in shades of transparent blue. Inset: Detail view of structural elements in the nucleotide-binding pocket before and after ADP release. (C) Cryo-EM map and corresponding atomic model of actin’s D-loop in ADP wild-type actomyosin-15. (D) Backbone positioning of In D-loop conformation in the indicated models, superimposed on the segmented cryo-EM map of the ADP wild-type actomyosin-15 D-loop.

• 

  Fig. 7.. ADP bound myosin-15 blunts F-actin nucleation. (A) Representative pyrene actin polymerization assays in the presence and absence of myosin-15 and ADP. Actin, 2 μM; myosin-15, 1 μM; Mg-ADP, 100 μM. A.U., arbitrary units. (B) Quantification of the time to half-maximal pyrene signal saturation in the presence of myosin-15 with or without ADP. Error bars represent SD. n = 5 from three independent protein purifications; **P = 0.008, Mann-Whitney U test. (C) Snapshots after 180 s of actin polymerization from TIRF movies recorded in the presence and absence of myosin-15 and ADP. Scale bar, 10 μm. (D) Quantification of F-actin density, indicative of nucleation rate, from TIRF movies. Solid lines and dashed lines/shading represent means ± SD at each time point; n = 3. (E) Representative kymographs of elongating individual actin filaments from TIRF movies. (F) Quantification of filament elongation rates from TIRF movies. n = 3 independent replicates. Bars represent mean; ****P < 0.0001, one-way analysis of variance. n.s., not significant.

• 

  Fig. 8.. Schematic model for regulation of actin polymerization by myosin-15. We propose myosin-15 enhances F-actin formation by stimulating nucleation, both by stabilizing actin-actin contacts and modulating D-loop structural plasticity. Myosin-15 nucleotide state facilitates tuning of nucleation activity, which could be harnessed for stereocilium height control. The jordan mutation disrupts F-actin regulation, thereby inhibiting polymerization.