Zebrafish Cardiac Muscle Thick Filaments: Isolation Technique and Three-Dimensional Structure

To understand how mutations in thick filament proteins such as cardiac myosin binding protein-C or titin, cause familial hypertrophic cardiomyopathies, it is important to determine the structure of the cardiac thick filament. Techniques for the genetic manipulation of the zebrafish are well established and it has become a major model for the study of the cardiovascular system. Our goal is to develop zebrafish as an alternative system to the mammalian heart model for the study of the structure of the cardiac thick filaments and the proteins that form it. We have successfully isolated thick filaments from zebrafish cardiac muscle, using a procedure similar to those for mammalian heart, and analyzed their structure by negative-staining and electron microscopy. The isolated filaments appear well ordered with the characteristic 42.9 nm quasi-helical repeat of the myosin heads expected from x-ray diffraction. We have performed single particle image analysis on the collected electron microscopy images for the C-zone region of these filaments and obtained a three-dimensional reconstruction at 3.5 nm resolution. This reconstruction reveals structure similar to the mammalian thick filament, and demonstrates that zebrafish may provide a useful model for the study of the changes in the cardiac thick filament associated with disease processes.     

Zebrafish Cardiac Muscle Thick Filaments: Isolation Technique and Three-Dimensional Structure

To understand how mutations in thick filament proteins such as cardiac myosin binding protein-C or titin, cause familial hypertrophic cardiomyopathies, it is important to determine the structure of the cardiac thick filament. Techniques for the genetic manipulation of the zebrafish are well established and it has become a major model for the study of the cardiovascular system. Our goal is to develop zebrafish as an alternative system to the mammalian heart model for the study of the structure of the cardiac thick filaments and the proteins that form it. We have successfully isolated thick filaments from zebrafish cardiac muscle, using a procedure similar to those for mammalian heart, and analyzed their structure by negative-staining and electron microscopy. The isolated filaments appear well ordered with the characteristic 42.9 nm quasi-helical repeat of the myosin heads expected from x-ray diffraction. We have performed single particle image analysis on the collected electron microscopy images for the C-zone region of these filaments and obtained a three-dimensional reconstruction at 3.5 nm resolution. This reconstruction reveals structure similar to the mammalian thick filament, and demonstrates that zebrafish may provide a useful model for the study of the changes in the cardiac thick filament associated with disease processes.     

Distribution of mass within native thick filaments of vertebrate skeletal muscle

The distribution of mass within the vertebrate skeletal thick filament has been determined by scanning transmission electron microscopy. Thick and thin filaments from fresh rabbit muscle were mixed with tobacco mosaic virus (TMV), fixed with formaldehyde, dried onto thin carbon films and viewed in a computer-linked microscope. Electron scattering data from both TMV and thick filaments were analysed with reference to the long axis of the particles so that the distribution of mass within the particles could be determined. While TMV appeared to be a uniform rod at the resolution employed (4.3 nm), the thick filament was clearly differentiated along its length. M-line remnants at the centre of the filament were flanked by regions of low mass per unit length, corresponding to the bare zone of the filament, and then by the more massive cross-bridge regions. The mass per unit length was approximately constant through most of the cross-bridge zone and declined at the filament tips, in a manner consistent with a constant number of myosin molecules per 14.3 nm interval (crown) throughout the cross-bridge zone. Fourier analysis of the data failed to detect the expected 43 nm periodicity of C-protein. The total mass of the thick filament was 184 Mdalton (s.e.m., 1.6 X 10(6); n = 70). The mass of adhering M-line proteins was highly variable but, on average, was about 4 Mdalton. The total mass of the filament and the mass distribution in the cross-bridge zone are consistent with three myosin molecules per crown.     

Axial arrangement of the myosin rod in vertebrate thick filaments: immunoelectron microscopy with a monoclonal antibody to light meromyosin

A monoclonal antibody, MF20, which has been shown previously to bind the myosin heavy chain of vertebrate striated muscle, has been proven to bind the light meromyosin (LMM) fragment by solid phase radioimmune assay with alpha-chymotryptic digests of purified myosin. Epitope mapping by electron microscopy of rotary-shadowed, myosin-antibody complexes has localized the antibody binding site to LMM at a point approximately 92 nm from the C-terminus of the myosin heavy chain. Since this epitope in native thick filaments is accessible to monoclonal antibodies, we used this antibody as a high affinity ligand to analyze the packing of LMM along the backbone of the thick filament. By immunofluorescence microscopy, MF20 was shown to bind along the entire A-band of chicken pectoralis myofibrils, although the epitope accessibility was greater near the ends than at the center of the A-bands. Thin-section, transmission electron microscopy of myofibrils decorated with MF20 revealed 50 regularly spaced, cross-striations in each half A-band, with a repeat distance of approximately 13 nm. These were numbered consecutively, 1-50, from the A-band to the last stripe, approximately 68 nm from the filament tips. These same striations could be visualized by negative staining of native thick filaments labeled with MF20. All 50 striations were of a consecutive, uninterrupted repeat which approximated the 14-15-nm axial translation of cross-bridges. Each half M-region contained five MF20 striations (approximately 13 nm apart) with a distance between stripes 1 and 1', on each half of the bare zone, of approximately 18 nm. This is compatible with a packing model with full, antiparallel overlap of the myosin rods in the bare zone region. Differences in the spacings measured with negatively stained myofilaments and thin-sectioned myofibrils have been shown to arise from specimen shrinkage in the fixed and embedded preparations. These observations provide strong support for Huxley's original proposal for myosin packing in thick filaments of vertebrate muscle (Huxley, H. E., 1963, J. Mol. Biol., 7:281-308) and, for the first time, directly demonstrate that the 14-15-nm axial translation of LMM in the thick filament backbone corresponds to the cross-bridge repeat detected with x-ray diffraction of living muscle.     

Effects of phosphorylation by myosin light chain kinase on the structure of Limulus thick filaments

The results discussed in the preceding paper (Levine, R. J. C., J. L. Woodhead, and H. A. King. 1991. J. Cell Biol. 113:563-572.) indicate that A-band shortening in Limulus muscle is a thick filament response to activation that occurs largely by fragmentation of filament ends. To assess the effect of biochemical changes directly associated with activation on the length and structure of thick filaments from Limulus telson muscle, a dually regulated tissue (Lehman, W., J. Kendrick-Jones, and A. G. Szent Gyorgyi. 1973. Cold Spring Harbor Symp. Quant. Biol. 37:319-330.) we have examined the thick filament response to phosphorylation of myosin regulatory light chains. In agreement with the previous work of J. Sellers (1981. J. Biol. Chem. 256:9274-9278), Limulus myosin, incubated with partially purified chicken gizzard myosin light chain kinase (MLCK) and [gamma 32P]-ATP, binds 2 mol phosphate/mole protein. On autoradiographs of SDS-PAGE, the label is restricted to the two regulatory light chains, LC1 and LC2. Incubation of long (greater than or equal to 4.0 microns) thick filaments, separated from Limulus telson muscle under relaxing conditions, with either intact MLCK in the presence of Ca2+ and calmodulin, or Ca2(+)-independent MLCK obtained by brief chymotryptic digestion (Walsh, M. P., R. Dabrowska, S. Hinkins, and D. J. Hartshorne. 1982. Biochemistry. 21:1919-1925), causes significant changes in their structure. These include: disordering of the helical surface arrangement of myosin heads as they move away from the filament backbone; the presence of distal bends and breaks, with loss of some surface myosin molecules, in each polar filament half; and the production of shorter filaments and end-fragments. The latter structures are similar to those produced by Ca2(+)-activation of skinned fibers (Levine, R. J. C., J. L. Woodhead, and H. A. King. J. Cell Biol. 113:563-572). Rinsing experimental filament preparations with relaxing solution before staining restores some degree of order of the helical surface array, but not filament length. We propose that outward movement of myosin heads and thick filament shortening in Limulus muscle are responses to activation that are dependent on phosphorylation of regulatory myosin light chains. Filament shortening may be due, in large part, to breakage at the filament ends.     

Electron microscopic study of troponin

Troponin and its components or fragments were observed in an electron microscope by the use of the rotary shadowing technique. In freshly prepared troponin with low viscosity, globular particles were mainly observed. The size of the long axis of the particles was 13.2 +/- 1.3 nm and the size perpendicular to the long axis was 9.5 +/- 1.2 nm. The mean axial ratio was 1.4 +/- 0.3. Most of the particles observed in a stored troponin preparation, having a higher viscosity than that of fresh troponin, had a globular head with a thin tail, with the total length of 25.4 +/- 1.4 nm (head-tail type particles). The axial size of the globular portion was 8.3 +/- 1.2 nm and the tail length was 17.1 +/- 1.6 nm. Observation of various particles during the transitional stages indicated that, in the globular particles, the tail region of head-tail type particle was associated along the globular head region. Troponin T was a filamentous particle with 16.9 +/- 1.5 nm length. The 26K fragment of troponin T, which was devoid of the N-terminal 45 residues from troponin T, was a filamentous particle with the length of 14.4 +/- 1.3 nm. Troponin T1, one of two chymotryptic subfragments of troponin T, was a filamentous particle of 11.6 +/- 1.4 nm length. Troponin C.T in the presence of Ca2+ was a particle with a globular head (7 nm in size) and a tail of about 17 nm length. The Fab fragment of anti-troponin T1 formed regular transverse striations along the thin filament of rabbit skeletal muscle with a 38 nm period.(ABSTRACT TRUNCATED AT 250 WORDS)     

The mammalian cardiac muscle thick filament: backbone contributions to meridional reflections

Information about the structure of the vertebrate striated muscle thick filament backbone is important for understanding the arrangement of both the rod portion of the myosin molecule and the accessory proteins associated with the backbone region of the filament. Although models of the backbone have been proposed, direct data on the structure of the backbone is limited. In this study, we provide evidence that electron micrographs of isolated negatively stained cardiac thick filaments contain significant information about the filament backbone. Computed Fourier transforms from isolated cardiac thick filaments show meridional (or near meridional) reflections on the 10th and 11th layer lines that are particularly strong. Comparison of Fourier filtrations of the filaments that exclude, or include, these reflections, provide evidence that these reflections originate at least in part from a series of striations on the backbone at a approximately 4 nm spacing. The striations are likely to result either from the packing of the myosin rods, or from proteins such as titin associated with the filament backbone.     

Lateral forces in the filament lattice of vertebrate striated muscle in the rigor state.

The repulsive pressure between filaments in the lattice of skinned rabbit and frog striated muscle in rigor has been measured as a function of interfilament spacing, using the osmotic pressure generated by solutions of large, uncharged polymeric molecules (dextran and polyvinylpyrrolidone). The pressure/spacing measurements have been compared with theoretically derived curves for electrostatic pressure. In both muscles, the major part of the experimental curves (100-2,000 torr) lies in the same region as the electrostatic pressure curves, providing that a thick filament charge diameter of approximately 30 nm in rabbit and approximately 26 nm in frog is assumed. In chemically skinned or glycerol-extracted rabbit muscle the fit is good; in chemically skinned frog sartorius and semitendinosus muscle the fit is poor, particularly at lower pressures where a greater spacing is observed than expected on theoretical grounds. The charge diameter is much larger than the generally accepted value for thick filament backbone diameter. This may be because electron microscope results have underestimated the amount of filament shrinkage during sample preparation, or because most of the filament charge is located at some distance from the backbone surface, e.g., on HMM-S2. Decreasing the ionic strength of the external solution, changing the pH, and varying the sarcomere length all give pressure/spacing changes similar to those expected from electrostatic pressure calculations. We conclude that over most of the external pressure range studied, repulsive pressure in the lattice is predominantly electrostatic.     

Influence of calcium on myosin thick filament formation in intact airway smooth muscle

Myosin thick filaments have been shown tobe structurally labile in intact smooth muscles. Although the mechanismof thick filament assembly/disassembly for purified myosins in solution has been well described, regulation of thick filament formation inintact muscle is still poorly understood. The present study investigates the effect of resting calcium level on thick filament maintenance in intact airway smooth muscle and on thick filament formation during activation. Cross-sectional density of the thick filaments measured electron microscopically showed that the density increased substantially (144%) when the muscle was activated. Theabundance of filamentous myosins in relaxed muscle was calcium sensitive; in the absence of calcium (with EGTA), the filament densitydeceased by 35%. Length oscillation imposed on the muscle underzero-calcium conditions produced no further reduction in the density.Isometric force and filament density recovered fully after reincubationof the muscle in normal physiological saline. The results suggest thatin airway smooth muscle, filamentous myosins exist in equilibrium withmonomeric myosins; muscle activation favors filament formation, and theresting calcium level is crucial for preservation of the filaments inthe relaxed state.

     

Myosin and paramyosin are organized about a newly identified core structure

Myosin isoforms A and B are differentially localized to the central and polar regions, respectively, of thick filaments in body wall muscle cells of Caenorhabditis elegans (Miller, D. M. III, I. Ortiz, G. C. Berliner, and H. F. Epstein, 1983, Cell, 34:477-490). Biochemical and electron microscope studies of KCl-dissociated filaments show that the myosin isoforms occupy a surface domain, paramyosin constitutes an intermediate domain, and a newly identified core structure exists. The diameters of the thick filaments vary significantly from 33.4 nm centrally to 14.0 nm near the ends. The latter value is comparable to the 15.2 nm diameter of the core structures. The internal density of the filament core appears solid medially and hollow at the poles. The differentiation of thick filament structure into supramolecular domains possessing specific substructures of characteristic stabilities suggests a sequential mode for thick filament assembly. In this model, the two myosin isoforms have distinct roles in assembly. The behavior of the myosins, including nucleation of assembly and determination of filament length, depend upon paramyosin and the core structure as well as their intrinsic molecular properties.     

Mutations in Drosophila myosin rod cause defects in myofibril assembly

The roles of myosin during muscle contraction are well studied, but how different domains of this protein are involved in myofibril assembly in vivo is far less understood. The indirect flight muscles (IFMs) of Drosophila melanogaster provide a good model for understanding muscle development and function in vivo. We show that two missense mutations in the rod region of the myosin heavy-chain gene, Mhc, give rise to IFM defects and abnormal myofibrils. These defects likely result from thick filament abnormalities that manifest during early sarcomere development or later by hypercontraction. The thick filament defects are accompanied by marked reduction in accumulation of flightin, a myosin binding protein, and its phosphorylated forms, which are required to stabilise thick filaments. We investigated with purified rod fragments whether the mutations affect the coiled-coil structure, rod aggregate size or rod stability. No significant changes in these parameters were detected, except for rod thermodynamic stability in one mutation. Molecular dynamics simulations suggest that these mutations may produce localised rod instabilities. We conclude that the aberrant myofibrils are a result of thick filament defects, but that these in vivo effects cannot be detected in vitro using the biophysical techniques employed. The in vivo investigation of these mutant phenotypes in IFM development and function provides a useful platform for studying myosin rod and thick filament formation generically, with application to the aetiology of human myosin rod myopathies.     

Structural analysis of viral replicative intermediates isolated from adenovirus type 2-infected HeLa cell nuclei.

Deoxyribonucleoprotein complexes released 17 h postinfection from adenovirus type 1 (Ad2)-infected HeLa cell nuclei were shown by electron microscopy to contain filaments much thicker (about 200 A [20 nm]) than double-stranded DNA (about 20 A [2 nm]). The complexes were partially purified through a linear sucrose gradient, concentrated, and further purified in a metrizamide gradient. The major protein present in the complexes was identified as the 72,000-dalton (72K), adenovirus-coded single-stranded DNA-binding protein (72K DBP). Three types of complexes have been visualized by electron microscopy. Some linear complexes were uniformly thick, and their length corresponded roughly to that of the adenovirus genome. Other linear genome-length complexes appeared to consist of a thick filament connected to a thinner filament with the diameter of double-stranded DNA. Forked complexes consisting of one thick filament connected to a genome-length, thinner double-stranded DNA filament were also visualized. Both thick and thin filaments were sensitive to DNase and not to RNase, but only the thick filaments were digested by the single-strand-specific Neurospora crassa nuclease, indicating that they correspond to a complex of 72K DBP and Ad2 single-stranded DNA. Experiments with anti-72K DBP immunoglobulins indicated that these nucleoprotein complexes, containing the 72K DBP, correspond to replicative intermediates. Both strands of the Ad2 genome were found associated to the 72K DBP. Altogether, our results establish the in vivo association of the 72K DBP with adenovirus single-stranded DNA, as previously suggested from in vitro studies, and support a strand displacement mechanism for Ad2 DNA replication, in which both strands can be displaced. In addition, our results indicate that, late in infection, histones are not bound to adenovirus DNA in the form of a nucleosomal chromatine-like structure.     

Protein osmotic pressure and cross-bridge attachment determine the stiffness of thin filaments in muscle ex vivo

The properties of some models of the actin filament are compared with those of the thin filament in muscle. The greater stiffness of thin filaments ex vivo with respect to F-actin in vitro is attributed to the effect of both protein osmotic pressure and the attached cross-bridges. By comparing the stiffness of thin filaments in vitro and in isometric and rigor muscles the stiffness of thin filaments in relaxed muscle is computed. The upper limit of thin filament stretching is deduced to approach approximately 10 nm microm(-1). It is also calculated that, on stretching by 2.02 nm of the fully non-overlapped thin filament or by 1.59 nm of the thin filament on isometric contraction, the energy released on the hydrolysis of one molecule of ATP is fully used up.     

Understanding the organisation and role of myosin binding protein C in normal striated muscle by comparison with MyBP-C knockout cardiac muscle

Myosin binding protein C (MyBP-C) is a component of the thick filament of striated muscle. The importance of this protein is revealed by recent evidence that mutations in the cardiac gene are a major cause of familial hypertrophic cardiomyopathy. Here we investigate the distribution of MyBP-C in the A-bands of cardiac and skeletal muscles and compare this to the A-band structure in cardiac muscle of MyBP-C-deficient mice. We have used a novel averaging technique to obtain the axial density distribution of A-bands in electron micrographs of well-preserved specimens. We show that cardiac and skeletal A-bands are very similar, with a length of 1.58 ± 0.01 μm. In normal cardiac and skeletal muscle, the distributions are very similar, showing clearly the series of 11 prominent accessory protein stripes in each half of the A-band spaced axially at 43-nm intervals and starting at the edge of the bare zone. We show by antibody labelling that in cardiac muscle the distal nine stripes are the location of MyBP-C. These stripes are considerably suppressed in the knockout mouse hearts as expected. Myosin heads on the surface of the thick filament in relaxed muscle are thought to be arranged in a three-stranded quasi-helix with a mean 14.3-nm axial cross bridge spacing and a 43 nm helix repeat. Extra “forbidden” meridional reflections, at orders of 43 nm, in X-ray diffraction patterns of muscle have been interpreted as due to an axial perturbation of some levels of myosin heads. However, in the MyBP-C-deficient hearts these extra meridional reflections are weak or absent, suggesting that they are due to MyBP-C itself or to MyBP-C in combination with a head perturbation brought about by the presence of MyBP-C.     

Calmodulin inhibits interaction of actin with MAP2 and Tau, two major microtubule-associated proteins

We have previously shown that microtubule-associated protein 2 (MAP2) and Tau, two major microtubule-associated proteins, interact with actin differently as measured by low-shear viscosity and that their activities are modified by phosphorylation (Nishida, E., Kotani, S., Kuwaki, T., and Sakai, H. (1982 in Biological Functions of Microtubules and Related Structures (Sakai, H., Mohri, H., and Borisy, G. G., eds) pp. 297-309, Academic Press, Japan). In the present study we further examined their interaction using turbidimetry, electron microscopy, low- and high-shear viscometry. MAP2 increased the low-shear viscosity of actin filament but had weaker effect on high-shear viscosity and turbidity of actin filaments. In contrast, Tau reduced high-shear viscosity of actin filaments and enhanced the turbidity which were due to formation of actin filament bundles as shown by electron microscopy. We conclude that MAP2 is a gelation factor, while Tau is a bundling factor. A well-known Ca2+-dependent regulatory protein, calmodulin, inhibited both MAP2-actin and Tau-actin interaction in a Ca2+-dependent manner. The calmodulin-dependent inhibition was canceled by higher concentrations of MAP2 or Tau, and calmodulin had no effect on the viscosity of actin filament alone, indicating that this inhibition is based on the stoichiometric interaction of calmodulin with MAP2 or Tau.     

The structure of thick filaments on longitudinal sections of rabbit psoas muscle

By means of electron microscopy the longitudinal sections of chemically skinned fibres of rigorised rabbit psoas muscle have been examined at pH of rigorising solutions equal to 6, 7, 8 (I = 0.125) and ionic strengths equal to 0.04, 0.125, 0.34 (pH 7.0). It has been revealed that at pH 6.0 the bands of minor proteins localization in A-disks were seen very distinctly, while at pH 7.0 and I = 0.125 these bands can be revealed only by means of antibody labelling technique. At the ionic strength of 0.34 (pH 7.0) the periodicity of 14.3 nm in thick filaments was clearly observed, which was determined by packing of the myosin rods into the filament shaft and of the myosin heads (cross-bridges) on the filament surface. The number of cross-bridge rows in the filament equals 102. A new scheme of myosin cross-bridge distribution in thick filaments of rabbit psoas muscle has been suggested according to which two rows of cross-bridges at each end of a thick filament are absent. The filament length equals 1.64 +/- 0.01 micron. It has been shown that the length of thick filament as well as the structural organization of their end regions in rabbit psoas muscle and frog sartorius one are different.     

An 100-kDa Ca2+-sensitive actin-fragmenting protein from unfertilized sea urchin egg

An actin-modulating protein was purified from unfertilized eggs of sea urchin, Hemicentrotus pulcherrimus, by means of DNase I affinity and DEAE-cellulose column chromatographies. This protein was a globular protein with a Stokes radius of 41-42 nm and consisted of a single polypeptide chain having an apparent molecular mass of 100 kDa on sodium dodecyl sulfate polyacrylamide gel electrophoresis. Gel filtration chromatography revealed that one 100-kDa protein molecule binds two or three actin monomers in the presence of Ca2+, but such binding was not observed in the absence of Ca2+. The effect of the 100-kDa protein on the polymerization of actin was studied by viscometry, spectrophotometry and electron microscopy. The initial rate of actin polymerization was decreased at a very low molar ratio of 100-kDa protein/actin. Acceleration of the initial rate of polymerization occurred at a relatively high, but still substoichiometric, molar ratio of 100-kDa protein/actin. The 100-kDa protein produced fragmentation of muscle actin filaments at Ca2+ concentrations greater than 0.3 microM as revealed by viscometry and electron microscopy. Evidence was also presented that the 100-kDa protein binds to the barbed end of the actin filament.     

The detection of smooth muscle desmin-like protein in BHK21/C13 fibroblasts.

Three distinct proteins, actin (42,000 daltons), the principal form of fibroblast 10 nm filament protein (55,000 daltons), and a protein with a molecular weight of 52,000 and a pI of 5.8 were detected in nonionic detergent-insoluble cytoskeletal and 10 nm filament preparations of control BHK21/C13 and line 9 hamster fibroblasts. Cytoskeletal preparations of other hamster fibroblast cell types, such as NIL8 and primary embryo fibroblasts, contained the 55,000-dalton component but lacked the 52,000-dalton protein. A Rous sarcoma virus transformant of the BHK21/C13 line and an adenovirus transformant of line 9, resembled the NIL8 and other fibroblast types in that they contained only the 55,000- and 42,000-dalton polypeptides. The identity of the 52,000-dalton protein in BHK21/C13 cells was studied. This protein co-electrophoreses on both one- and two-dimensional polyacrylamide gels with the predominant muscle form of 10 nm filament protein. Further, one-dimensional peptide maps of the hamster smooth muscle 10 nm filament protein and the hamster fibroblast 52,000-dalton protein are identical to one another and distinct from the peptide maps of both the 42,000- and the 55,000-dalton components of the fibroblast cytoskeletal preparations. We conclude that BHK21/C13 cells contain both the fibroblast and the muscle form of 10 nm filament protein.     

Purified thick filaments from the nematode Caenorhabditis elegans: evidence for multiple proteins associated with core structures

The thick filaments of the nematode, Caenorhabditis elegans, arising predominantly from the body-wall muscles, contain two myosin isoforms and paramyosin as their major proteins. The two myosins are located in distinct regions of the surfaces, while paramyosin is located within the backbones of the filaments. Tubular structures constitute the cores of the polar regions, and electron-dense material is present in the cores of the central regions (Epstein, H.F., D.M. Miller, I. Ortiz, and G.C. Berliner. 1985. J. Cell Biol. 100:904-915). Biochemical, genetic, and immunological experiments indicate that the two myosins and paramyosin are not necessary core components (Epstein, H.F., I. Ortiz, and L.A. Traeger Mackinnon. 1986. J. Cell Biol. 103:985-993). The existence of the core structures suggests, therefore, that additional proteins may be associated with thick filaments in C. elegans. To biochemically detect minor associated proteins, a new procedure for the isolation of thick filaments of high purity and structural preservation has been developed. The final step, glycerol gradient centrifugation, yielded fractions that are contaminated by, at most, 1-2% with actin, tropomyosin, or ribosome-associated proteins on the basis of Coomassie Blue staining and electron microscopy. Silver staining and radioautography of gel electrophoretograms of unlabeled and 35S-labeled proteins, respectively, revealed at least 10 additional bands that cosedimented with thick filaments in glycerol gradients. Core structures prepared from wild-type thick filaments contained at least six of these thick filament-associated protein bands. The six proteins also cosedimented with thick filaments purified by gradient centrifugation from CB190 mutants lacking myosin heavy chain B and from CB1214 mutants lacking paramyosin. For these reasons, we propose that the six associated proteins are potential candidates for putative components of core structures in the thick filaments of body-wall muscles of C. elegans.