Myosin head configuration in relaxed insect flight muscle: x-ray modeled resting cross-bridges in a pre-powerstroke state are poised for actin binding

Abstract

Low-angle x-ray diffraction patterns from relaxed insect flight muscle recorded on the BioCAT beamline at the Argonne APS have been modeled to 6.5 nm resolution (R-factor 9.7%, 65 reflections) using the known myosin head atomic coordinates, a hinge between the motor (catalytic) domain and the light chain-binding (neck) region (lever arm), together with a simulated annealing procedure. The best head conformation angles around the hinge gave a head shape that was close to that typical of relaxed M*ADP*Pi heads, a head shape never before demonstrated in intact muscle. The best packing constrained the eight heads per crown within a compact crown shelf projecting at approximately 90 degrees to the filament axis. The two heads of each myosin molecule assume nonequivalent positions, one head projecting outward while the other curves round the thick filament surface to nose against the proximal neck of the projecting head of the neighboring molecule. The projecting heads immediately suggest a possible cross-bridge cycle. The relaxed projecting head, oriented almost as needed for actin attachment, will attach, then release Pi followed by ADP, as the lever arm with a purely axial change in tilt drives approximately 10 nm of actin filament sliding on the way to the nucleotide-free limit of its working stroke. The overall arrangement appears well designed to support precision cycling for the myogenic oscillatory mode of contraction with its enhanced stretch-activation response used in flight by insects equipped with asynchronous fibrillar flight muscles.

FIGURE 1

EM images and analysis of a myac layer (myosin and actin alternate where a 25 nm section includes single filament layer) from glycerinated, unstretched, Lethocerus IFM, plunge-frozen to −190°C in the relaxed state (5 mM Mg-ATP), then freeze-substituted in acetone at −80°C via TAURAC fixation. Scale bars = 232 nm. (a) EM shows ∼70% of length of a typical 2.67 μm sarcomere. Thick filaments keep a tight lateral register across the A-band of axial 14.5-nm cross-bridge repeat, despite a loose whole-filament register (meander of Z- and M-bands). (b) The same preparation as a but a different region showing clear C-filament connections to the Z-band. (c and d) The filtered image brings out a long 116-nm repeat as “beating” of a 14.5-nm myosin repeat against two 38.7-nm pseudorepeats of actin, one intrinsic to thick filaments, the other to thin filaments. In d, 38.7 nm appears as denser cross-bridge contacts with dense segments (troponin) along thin filaments, in contrast to rigor and active contraction where cross-bridges between troponins are predominant (Taylor et al., 1999). (e) The computed image transform from the A-band region between arrowheads in a shows that cryofixation has preserved native ordering of layer-line relative intensities (14.5 > 38.7 > 23.2 > 19.3 nm; see native x-ray pattern of Fig. 2 a). By contrast, direct chemical fixation suppresses 23.2 nm and enhances 19.3 nm (giving 14.5 > 38.7 > 19.3 ≫ 23.2 nm) (Reedy et al., 1983, 1987). Scale bars in a, b, and d all show ∼16 repeats of the 14.5-nm repeat of myosin head crowns.

FIGURE 2

(a) The full relaxed insect low-angle x-ray diffraction pattern, showing 20 layer-lines, based on a 116-nm repeat, and ∼468 Bragg reflections. This pattern is obtained from relaxed bundles of glycerinated flight muscle from the waterbug Lethocerus indicus in the Mg-ATP relaxed state. (b) Surface helical net for vertebrate skeletal and IFM thick filaments comparing the helical repeat and the true axial repeat of the two structures. Rotation from crown to crown is 40° in vertebrate skeletal muscle and 33.75° in IFM. Vertebrate skeletal muscle has threefold rotational symmetry and IFM has fourfold. (c) Unit cell in transverse view for IFM, frog, and fish (from left to right, respectively) and their lattice spacings (A is actin filament, M is myosin filament). IFM and fish have a simple lattice, whereas frog has a superlattice with the myosin filaments not in rotational register.

FIGURE 3

(a) The myosin head structure (Rayment et al., 1993a) as downloaded from the pdb database showing the definitions of the head orientation in terms of the x, y, and z axes and illustrating the myosin heavy chain and the two light chains (the regulatory light chain (green) and essential light chain (blue)). The long α–helical part of the heavy chain in red in the core of the regulatory light chain region provides the link between the heads and the rest of the myosin filament. The actin-binding face of the myosin head is top right and slightly facing up and out of the page. The motor or catalytic domain and the neck or lever arm are indicated. (b) Representation of a in terms of a 59-sphere model. The volume of these spheres, each of radius 8.61 Å, was chosen to express the overall mass of the myosin head, assuming constant protein density within all spheres. The black dots in a and b represent the position of the origin (0, 0, 0). (c–h) Three different views of the head in a and of its simulation in b illustrating the definition of the three head reference axes x, y, and z. These are initially set parallel to muscle reference axes X, Y, and Z. The parameters discussed in the text and in Fig. 4 refer to movements around these axes. The z axis is parallel to the muscle long axis which we call Z. If the x axis starts pointing out radially from the filament surface, this defines the filament axis X. The Y axis is orthogonal to X and Z. The head tilt (β) refers to the tilt of the head x axis around the Y-reference axis in the X-Z plane. The slew (α) refers to the rotation of the head x axis around the muscle Z axis in the X-Y plane. The rotation (θ) refers to movements around the head x axis. a shows the zero tilt position of the head, but clearly the head mass is already tilted up. The axial range of tilts was therefore set at 45° upward and 135° downward since beyond these values the heads would sterically clash with the backbone. (f–h) These show the head orientations as in c–e but for the representation of the head as a set of 59 spheres. The initial searches were carried out with the 59-sphere model (f–h). Final searches and refinements were carried out with the full pdb structure (c–e).

FIGURE 4

Illustration of the 13 model parameters that were used in the searches and which are detailed in the caption to Fig. 3 and in Table 2.

FIGURE 5

(a) Lower right quadrant of the observed relaxed IFM x-ray diffraction pattern stripped using CCP13 analysis and used for modeling 65 reflections with a resolution cut-off of 6.5 nm. The myosin layer-line spacings are shown at the left. (b) The diffraction produced from the best model (illustrated in d). (c) The calculated diffraction pattern including the meridional reflections and the 38.6-nm layer-line, none of which were used in the modeling. Only the myosin contribution on the 38.6-nm layer-line is shown. The remainder of that layer-line is presumed to come from the actin filaments (outer end) and troponin (inner end). (d) Stereo image of the best model for the relaxed IFM myosin filament using the observed data in a. The heads were modeled at 20 Å resolution using the pdb atomic coordinates. Also, on the right (e), the best model is shown with the heads displayed as a 59-sphere model, with the white arrows showing the direction of the two heads in one myosin molecule and the red spot marking the shared origin of a corresponding pair of heads (also shown in d). The backbone is shown in d and e as a simple cylinder. It has been included only for visual clarity and has not been fitted to the data. The image in d toward the top shows the inner heads colored green and the outer heads colored pink. In e, heads in equivalent positions along helical tracks through different 14.5-nm-spaced crowns are the same color.

FIGURE 6

(a and b) The lowest R-factor structures using a Rayment head (left; R-factor 22.2%) and the new modeled head (right; R-factor 9.7%), both viewed looking down the filament axis toward the Z-band. In each case, a shows the relationship between the two heads of one myosin molecule and b shows one full crown of eight heads. Note that the Rayment structure (left) has sterically clashing heads. (c–g) Comparison of x-ray and electron microscopy models (Morris et al., 1991). c, d are projections of the best model of this study at 20 Å and 60 Å resolution, respectively, and e–g are projections of the IFM thick filament 3D reconstruction of Morris et al. (1991) based on analysis of electron micrograph images of negatively-stained insect thick filaments with the 3D map recalculated with 100%, 50%, and 0% weighting for the equator. Among other things, reduced weighting of the equator reduces the contribution from the backbone, perhaps making the EM model more comparable to the x-ray model. Comparison of c, d, and e–g suggests that the polarity of the x-ray model is such that the M-band is at the top, as it is known to be in the electron micrograph reconstructions. (h) Two views of a single myosin head crown in the best model showing the actin-binding sites on the myosin heads. As in a and b, the left image is a view looking Z-ward from the M-band and the right shows a side view with the actin filament axis nearly vertical, with the M-band at the top, Z-band at bottom.

FIGURE 7

Comparing S1 from our model with published crystal structures. All are superimposed at the catalytic domain, so that the differences between models are expressed by the different positions of the neck region or lever-arm. (a and b) Stereo views. Green and pointing slightly toward the viewer is the Rayment et al. (1993a, chicken skeletal myosin with no nucleotide (i.e., rigor-like); dark blue is the Dominguez et al. (1998) chicken smooth muscle myosin in ADP•AlF₄ form; pale blue is our Lethocerus model; orange is the Houdusse et al. (2000) scallop myosin in Mg•ADP•VO₄ form. The view in a has the actin filament axis vertical and to the right, with the M-band at the top and Z-band at the bottom, whereas b is the view down the actin filament axis toward the Z-band (behind the page). (c and d) Direct comparison of the relaxed Lethocerus head shape with the Rayment et al. (1993a) structure. Green is the Rayment rigor structure and pale blue shows our Lethocerus model superimposed on the Rayment structure (Z-band is at the bottom in c and behind the page in d).

FIGURE 8

(a) Defining the different motions of the lever arm and hook axis denoted by neck tilt, neck slew, and neck rotation in different myosin head structures, assuming that the catalytic domain is oriented as it would be if attached to actin in the rigor conformation, with the actin axis vertical and the Z-band end at the bottom. The neck long axis is taken as the line joining residues 780 and 843, and the hook axis is taken as the line joining residues 830 to 843. (b) A plot showing the neck tilt and neck slew angles required by the lever arm in order to convert between the four structures when the catalytic domain is superimposed on a vertical actin filament in the rigor attachment position (see Fig. 7, a and b). The arrows on different points show the relative neck rotations of the hook region of the myosin head in the different structures. The vertical reference axis is the actin filament axis with M-band up, Z-band down. R refers to the Rayment rigor structure (Rayment et al., 1993a,b), D refers to the Dominguez ‘prepowerstroke’ structure (Dominguez et al., 1998), L is Lethocerus (this work), and S is the scallop myosin Mg•ADP•VO₄ structure (Houdusse et al., 2000). (c and d) The relaxed to rigor transition involves a swing of the myosin neck domain. Our best, relaxed myosin thick filament model is shown on the left (c), and an actin filament reconstruction labeled with S1 in the rigor state is shown on the right (d) (Harford and Squire, unpublished data from S1-labeled fish muscle). For an actin target monomer ideally opposite a relaxed myosin head, the relaxed to actin-bound (prepowerstroke) transition requires only a radial movement of ∼2 nm and a small rotation of the catalytic domain to achieve rigor-like docking of the catalytic domain on the actin site (red arrow). After docking, only a purely axial swing (no slew), with a small pivot rotation (of +20°), of the myosin head neck domain (lever-arm) is needed to complete a powerstroke and to reach the final rigor conformation (as in Fig. 7, c and d). The lateral separation of the myosin and actin filaments has been exaggerated in this diagram.