Hugh E. Huxley

X-ray diffraction measurements of the extensibility of actin and myosin filaments in contracting muscle.

We have used a small angle scattering system assembled on the high flux multipole wiggler beam line at CHESS (Cornell) to make very accurate spacing measurements of certain meridional and layer-line reflections from contracting muscles. During isometric contraction, the actin 27.3 A reflection increases in spacing from its resting value by approximately 0.3%, and other actin reflections, including the 59 and 51 A off-meridional reflections, show corresponding changes in spacing. When tension is augmented or diminished by applying moderate speed length changes to a contracting muscle, changes in spacing in the range of 0.19-0.24% (when scaled to full isometric tension) can be seen. The larger difference between the resting and isometric spacings suggests either nonlinearity at low tension levels or the presence of a component related to activation itself. Myosin filaments also show similar increases in axial period during slow stretch, in addition to the well known larger change associated with activation. An actin spacing change of 0.25-0.3% can also be measured during a 2 ms time frame immediately after a quick release, showing that the elastic behavior is rapid. These observations of filament extensions totaling 2-3 nm per half-sarcomere may necessitate some significant revision of the interpretation of a number of mechanical experiments in muscle, in which it has usually been assumed that virtually all of the elasticity resides in the cross-bridges.

Electron microscope observations on the structure of microsomal particles from Escherichia coli

Microsomal particles from E. coli are known to exist in a number of different stable forms, depending foremost on the concentration of free magnesium ions, and characterized by sedimentation coefficients of 100 s, 70 s, 50 s and 30 s. These different forms, identified in each case by sedimentation analysis, have been investigated by electron microscopy, using both negative and positive staining. Using negative staining it was found that the 100 s particles consisted of two 70 s particles joined together as a dimer. The 70 s particles could be seen to contain two subunits of unequal size with a well-marked cleft between them, and in the 100 s particles the smaller subunits of the two 70 s particles were apposed. 50 s and 30 s particles were found to resemble closely the larger and the smaller sub-units of the 70 s particles. In addition, it was found that purified 50 s particles, when transferred to a medium in which 100 s particles were stable, formed dimers with a sedimentation coefficient of about 81s. The particles were not found to have a regular polyhedral form, in contrast to some small spherical viruses. The 50 s particles appeared to be very approximately dome-shaped. The 30 s particles had a flatter form with a rather irregular outline. In the 70 s particles, the smaller subunit was seen to fit like an overlapping cap on the flattened face of the larger one. When microsomal particles were stained with uranyl acetate (which in many situations appears to act as a preferential stain for nucleic acids) they showed a complex internal structure, but no indication of the cleft between the subunits in 70 s particles. No evidence was found for the existence of a protein shell around a nucleic acid core in the particles, again in contrast to some small spherical viruses.

Quantitative studies on the structure of cross-striated myofibrils: I. Investigations by interference microscopy

The quantity of A substance in glycerol-extracted fibrils of rabbit psoas muscle has been determined by interference microscopy and found to be 50–55%, most probably 54.5%, of the total protein of the fibril. When myosin-extracting procedures are applied to these fibrils, all of the A substance is removed, together with another 10% of the total protein, making a total of 60–65% extracted. In the accompanying paper12 it has been shown by chemical analysis of similar fibrils that myosin constitutes about 51% of the total protein of the fibril, and that the myosin-extracting procedures remove about 62% of the total protein. These comparative measurements by interference microscopy and chemical analysis prove that at least four-fifths of the myosin in these fibrils is present as the A substance, and the results are in excellent agreement with the hypothesis that all the myosin is concentrated in the A bands.

X-ray analysis and the problem of muscle.

The present-day picture of muscle is briefly as follows: muscle is a machine for converting chemical energy into mechanical work; the ‘moving parts’ of this machine are built up of two proteins, actin and myosin; the known energy-producing reaction most closely linked to the contractile process is the dephosphorylation of adenosine triphosphate (ATP). In the present studies, low-angle X-ray diffraction technique has been used to study the molecular structure of these two proteins, actin and myosin, and their arrangement in living muscle under various conditions.

Calcium sensitive binding of troponin to actin-tropomyosin: A two-site model for troponin action

Abstract Troponin reconstituted from the inhibitory component (troponin-I) and calcium binding protein (troponin-C) binds readily to actin-tropomyosin in 0.1 m m -EGTA but only poorly in 0.01 m m -CaCl 2 or 0.1 m m -Ca-EGTA. Troponin prepared by extraction of myofibrils with mersalyl, an organic mercurial, contains only these two components and also shows this calcium-sensitive binding and is deficient in its ability to bind to tropomyosin. Troponin-I + C is unable to confer calcium sensitivity on the Mg 2+ activated actomyosin ATPase in concentrations at which native troponin is fully effective and the ATPase activity remains high in the absence of calcium. Addition of the tropomyosin binding component (troponin-T) to the other two components restores their ability to remain associated with actin-tropomyosin in the presence of calcium as does native troponin; calcium sensitivity is also regained. The results of these experiments have been interpreted in terms of a two-site mechanism of troponin action.

Quantitative studies on the structure of cross-striated myofibrils: II. Investigations by biochemical techniques

The quantities of various protein fractions in the psoas muscle of the rabbit have been determined by chemical analysis. The soluble protein removed by glycerol extraction and by the subsequent isolation and washing of the fibrils in neutral hypotonic salt solution, make up 34% of the total protein of the whole muscle. 62% of the total protein of the washed fibrils is taken out by myosin-extracting procedures. 82% of the extracted protein precipitates when the ionic strength is reduced to 0.04; this component is conventionally known as myosin. Thus the myosin extracted makes up 51% of the total protein of the washed fibrils, and the removal of the myosin is accompanied by the extraction of a further 11% of the total protein. In the accompanying paper5 it has been shown by interference microscopy that the quantity of A substance in similar fibrils is 50–55% of their total protein, and that the same myosin-extracting procedures remove the A substance and another 10% of the total protein, 60–65% in all. These comparative measurements by interference microscopy and chemical analysis prove that at least four-fifths of the myosin in these fibrils is present as the A substance, and the results are in excellent agreement with the hypothesis that all the myosin is concentrated in the A bands.

The structure of the protein shell of turnip yellow mosaic virus

Turnip yellow mosaic virus (TYMV) is an approximately spherical plant virus, having a diameter of 280 to 300 A and containing 60% protein and 40% RNA; the protein is believed to form a spherical shell around a nucleic acid core. It has been shown previously by X-ray diffraction that the virus particles must have cubic symmetry, and it has been argued on theoretical grounds that viruses having cubic symmetry and containing a spherical shell of protein were likely to have that protein shell built up of subunits arranged in a polyhedron having either tetrahedral (23), octahedral (432) or icosahedral (532) symmetry elements. We have examined TYMV in the electron microscope by the negative staining method, and have found that regularly disposed subunits may be discerned on the surface of the particles. There appear to be 32 such subunits, arranged on or near the vertices of either a pentakis dodecahedron or a rhombic triacontahedron (or other polyhedra approximating to those forms). These are semi-regular poly-hedra having 532 symmetry, the former having 60 identical triangular faces and the latter 30 identical rhombic faces; both have 32 vertices, which fall into two separate groups, 12 vertices lying on fivefold axes and 20 vertices lying on threefold axes. Thus the subunits occupy two different types of position, and can be of two different species. Because the subunits lie on rotation axes, they must contain the appropriate number of identical structural units. These results agree with the suggestion made on the basis of the X-ray results that the particles have 532 symmetry, but they conflict with the more tentative suggestion that the particular polyhedron involved contained 60 subunits placed at the vertices of a snub dodecahedron. A possible way in which the X-ray results can be reconciled with the present observations is suggested.

Ultrastructure of skeletal muscle fibers studied by a plunge quick freezing method: myofilament lengths

We have set up a system to rapidly freeze muscle fibers during contraction to investigate by electron microscopy the ultrastructure of active muscles. Glycerinated fiber bundles of rabbit psoas muscles were frozen in conditions of rigor, relaxation, isometric contraction, and active shortening. Freezing was carried out by plunging the bundles into liquid ethane. The frozen bundles were then freeze-substituted, plastic-embedded, and sectioned for electron microscopic observation. X-ray diffraction patterns of the embedded bundles and optical diffraction patterns of the micrographs resemble the x-ray diffraction patterns of unfixed muscles, showing the ability of the method to preserve the muscle ultrastructure. In the optical diffraction patterns layer lines up to 1/5.9 nm-1 were observed. Using this method we have investigated the myofilament lengths and concluded that there are no major changes in length in either the actin or the myosin filaments under any of the conditions explored.

OCCURRENCE AND FUNCTION OF POLYSOMES IN RABBIT RETICULOCYTES.

Electron microscopy of sections of rabbit reticulocytes reveals that the ribosomes are arranged in groups or clusters. Those with four or five 80 s-ribosomes predominate, with the particles arranged as if on the circumference of a circle. Isolated ribosome preparations, viewed by positive or negative staining, show similar aggregates, or “polysomes”, although at times the ribosomes in isolated preparations appear to be arranged in a line and connected by a strand of densely stained material. On analytical or density-gradient centrifugation of cell lysates or isolated ribosome preparations, peaks of sedimentation coefficient from 80 s to 250 s are seen. Examination by electron microscopy of material from each peak demonstrates that they correspond to aggregates of a definite number of ribosomes. The ribosomes of the polysome are linked by a substance sensitive to very low concentrations of pancreatic ribonuclease, presumably an RNA molecule. Rapid incorporation of amino acid into polysomes was demonstrated. RNA, isolated by a lithium chloride technique, was tested for its ability to enhance incorporation of amino acid; the 16 s, 28 s and 41 s components all gave roughly equal enhancement. Assays were performed to measure the leucine aminopeptidase and ribonuclease activities associated with isolated ribosomes. Only very low levels of both enzymes were found.

Cross-bridge movement during muscle contraction.

WHEN vertebrate striated muscle contracts at constant length, the 1,0 reflection decreases in intensity while the 1,1 reflection increases in intensity1,12. These changes have been interpreted as arising from the movement of the cross bridges from the vicinity of the thick filaments in relaxed muscle to the vicinity of the thin filaments in the contracting state, and this movement has been thought to involve both radial and azimuthal components2,3. Lymn4 has claimed that models with cross bridges near the thick filament do not simulate the intensity distribution of the higher order equatorial reflections of relaxed muscle, while models in which the cross bridges are extended to near the radius of the thin filaments do give good simulations. This has led him to contest the interpretation of the intensity changes given above, and to propose instead a model in which these changes are brought about by azimuthal movement alone. In this article we challenge Lymns interpretation on several grounds.