Three-dimensional structure of the vertebrate muscle A-band: III. M-region structure and myosin filament symmetry

Abstract

Ultrathin transverse sections of the body muscle of bony fish and the sartorius muscle of frog have been analysed in detail by optical diffraction and image averaging to reveal the ultrastructures of the myosin filaments both in the M-region and in the outer ends of the filament (the tip region). The evidence is unequivocal that the myosin filaments in both muscle types have 3-fold rotational symmetry in all of the regions where symmetry can be seen. Using the nomenclature of Sjöström & Squire (1977 a), the appearances of cross-sections in fish at different axial locations are as follows. (Note that fish myosin filaments are arranged in a simple hexagonal lattice.) 1. (1)At M1 (central M-bridge line) the myosin filament profile is almost circular, but does sometimes show a three-component structure. There are six M-bridges from each filament. 2. (2) At M4 the filament backbone comprises three kidney-shaped subunits related by a triad axis, together with six M-bridges. There are two types of M-bridge interaction site. 3. (3) At M6 there are three kidney-shaped subunits related by a triad. There are no M-bridges but there does seem to be extra protein here. 4. (4) In the bare region there are three nearly circular subunits related by a triad. 5. (5) At M9 or the start of the bridge region there is a Y-shaped filament profile from which three projections radiate towards three of the six actin filament positions (trigonal points). 6. (6) There is an apparent change of hand of the filament profiles across the M-band when viewed in the same section (i.e. from M4 to M4′ or M6 to M6′). This implies that the bipolar myosin filament has the symmetry of the dihedral point group 32. 7. (7) There is a change in orientation of the 3-fold myosin filament profile across the M-band so that the bare region profiles are about 40 ° apart. Less detail has been seen in frog sartorius muscle sections because of the presence of the superlattice (see paper II in this series, Luther & Squire, 1980). However, the observations are entirely compatible with a myosin filament with the same structure as in fish but arranged in a superlattice. The superlattice is apparent not only in the bare region in frog, but also at the M-bridge level M4. There are two classes of M-bridge interaction at M4 just as in fish. At the A-band edges of both muscles (outer D-zone) the filament profiles appear either triangular, Y-shaped or composed of three subunits. The lattices are insufficiently ordered for satisfactory image averaging to be carried out. As yet the main part of the bridge region shows little ultrastructural detail. However, the present results show clearly that the vertebrate myosin filament has 3-fold rotational symmetry; there can be little doubt that the symmetry of the crossbridge array approximates to that of a three-stranded helix (Squire, 1972). A model is proposed for the structure of the myosin filament in the vertebrate M-region. It explains the observed appearances without going into molecular detail. It is also suggested that it is the M-bridges at M4 that are primarily responsible for defining the two types of A-band structure (simple lattice and superlattice) and that the M1 bridges may have a secondary role.