Atomic force microscopic evidence for Z-band as a rigid disc fixing the sarcomere structure of skeletal muscle

Atomic force microscopic images of single skeletal myofibrils showed periodical broad filamentous bands interspaced with narrow rigid bands corresponding to the sarcomere structures of skeletal muscle (Yoshikawa, Y., Yasuike, T., Yagi, A., and Yamada, T. 1999. Biochem. Biophys. Res. Comm., 256: 13-19). In order to identify the narrow rigid bands, comparative studies were made for intact single myofibrils and those treated with calcium-activated neutral protease by use of atomic force microscopy. It was found that (a) the periodical narrow rigid bands present in intact myofibrils were completely absent in myofibrils treated with calcium-activated neutral protease, and that (b) myofibrils treated with calcium-activated neutral protease were very fragile compared with intact myofibrils. As calcium-activated neutral protease selectively removes Z-bands of myofibrils (Reddy, M. K., Etlinger, J. D., Rabinowitz, M., Fischman, D. A., and Zak, R. 1975. J. Biol. Chem., 250: 4278-4284), these results clearly indicate that (a) the narrow rigid bands are the Z-bands, and that (b) the Z-bands are the essential disc supporting the sarcomere structure of skeletal muscle.     

The positional stability of thick filaments in activated skeletal muscle depends on sarcomere length: evidence for the role of titin filaments

Electron microscopy was used to study the positional stability of thick filaments in isometrically contracting skinned rabbit psoas muscle as a function of sarcomere length at 7 degrees C. After calcium activation at a sarcomere length of 2.6 micron, where resting stiffness is low, sarcomeres become nonuniform in length. The dispersion in sarcomere length is complete by the time maximum tension is reached. A-bands generally move from their central position and continue moving toward one of the Z-discs after tension has reached a plateau at its maximum level. The lengths of the thick and thin filaments remain constant during this movement. The extent of A-band movement during contraction depends on the final length of the individual sarcomere. After prolonged activation, all sarcomeres between 1.9 and 2.5 micron long exhibit A-bands that are adjacent to a Z-disc, with no intervening I-band. Sarcomeres 2.6 or 2.7 micron long exhibit a partial movement of A-bands. At longer sarcomere lengths, where the resting stiffness exceeds the slope of the active tension-length relation, the A-bands remain perfectly centered during contraction. Sarcomere symmetry and length uniformity are restored upon relaxation. These results indicate that the central position of the thick filaments in the resting sarcomere becomes unstable upon activation. In addition, they provide evidence that the elastic titin filaments, which join thick filaments to Z-discs, produce almost all of the resting tension in skinned rabbit psoas fibers and act to resist the movement of thick filaments away from the center of the sarcomere during contraction.     

Beta-actinin: a capping protein at the pointed end of thin filaments in skeletal muscle

We examined the function of beta-actinin as a pointed end capping protein of thin filaments in skeletal muscle. An improvement in preparing beta-actinin yielded purified beta-actinin which retained its activity for more than a week. Two-dimensional gel electrophoresis showed that the two subunits, beta I and beta II, of beta-actinin are, respectively, split into two to three components (isoforms) with different isoelectric points. Polyclonal antibody was raised by injecting such purified and undenatured chicken breast muscle beta-actinin composed of several components into a rabbit. Immuno-gold labeling examination with electron microscopy of an F-actin-beta-actinin complex decorated with HMM showed that 85% of bound gold particles was on the pointed end of actin filaments, while the remaining 15% was on the barbed end. This suggests that in beta-actinin preparation pointed end and barbed end capping proteins inevitably coexist. Immunofluorescence and immunoelectron microscopy directly showed that beta-actinin is located at the pointed end of thin filaments in myofibrils; it was also suggested that a capping protein having common antigenic determinants to beta-actinin is located at Z-line. Thus, the physiological function of beta-actinin as a pointed end capping protein was examined as follows: When beta-actinin was dissociated from the pointed end of thin filaments in an I-Z-I brush by using a high salt solution, thin filaments could be disassembled at the pointed ends at concentrations of exogenous actin lower than a critical value. At a physiological ionic strength, these salt-washed thin filaments gradually shortened at a constant rate of about 45 nm/h. Both the association and dissociation of monomeric actin at the pointed end were suppressed by the rebinding of exogenous beta-actinin. The main physiological role of beta-actinin is therefore to stabilize thin filaments in the sarcomere by preventing addition and removal of actin monomers at the pointed filament end.     

The organization of titin filaments in the half-sarcomere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line

mAbs specific for titin or nebulin were characterized by immunoblotting and fluorescence microscopy. Immunoelectron microscopy on relaxed chicken breast muscle revealed unique transverse striping patterns. Each of the 10 distinct titin antibodies provided a pair of delicate decoration lines per sarcomere. The position of these pairs was centrally symmetric to the M line and was antibody dependent. The results provided a linear epitope map, which starts at the Z line (antibody T20), covers five distinct positions along the I band (T21, T12, T4, T1, T11), the A-I junction (T3), and three distinct positions within the A band (T10, T22, T23). The epitope of T23 locates 0.2 micron before the M line. In immunoblots, the two antibodies decorating at or just before the Z line (T20, T21) specifically recognized the insoluble titin TI component but did not recognize TII, a proteolytic derivative. All other titin antibodies recognized TI and TII. Thus titin molecules appear as polar structures lacking over large regions repetitive epitopes. One physical end seems related to Z line anchorage, while the other may bind close to the M line. Titin epitopes influenced by the contractional state of the sarcomere locate between the N1 line and the A-I junction (T4, T1, T11). We discuss the results in relation to titin molecules having half-sarcomere lengths. The three nebulin antibodies so far characterized again give rise to distinct pairs of stripes. These locate close to the N2 line.     

Extra actin filaments at the periphery of skeletal muscle myofibrils.

Myofibrils isolated from a variety of vertebrate muscle fibers have a set of peripheral filaments associated with the periphery of the Z line free to move away from the surface of the myofibril. Decoration with myosin subfragment 1 shows that these are actin filaments.     

Titin: quantitative mass measurements by scanning transmission electron microscopy and structural implications for the sarcomere matrix of skeletal muscle

Scanning transmission electron microscopy has been used to investigate mass and linear mass density of native titin-2, a large soluble fragment of intact titin, from rabbit skeletal muscle. Dark field images of unstained, freeze-dried titin-2 appeared as either compact globules or looser and larger balls of string. Direct mass measurements indicated that the compact forms have an average mass of 2.40 +/- 0.50 x 10(6) Da. The mass to length ratio, determined from well-spread portions of titin strands (3-5 nm wide) from the ball of string forms, averaged 2.7 +/- 0.9 kDa/nm. Thus a single native intact titin molecule has a calculated contour length of well above approximately 1 micron, sufficient to span unidirectionally between the Z line and M line region in a resting-length sarcomere.     

The M-band: an elastic web that crosslinks thick filaments in the center of the sarcomere

The sarcomere of striated muscle is an efficient molecular machine, characterized by perfect structural organization of contractile filaments. This order is ensured by the sarcomere cytoskeleton, an important element of which is the M-band, believed to maintain the thick filament lattice. We review here recent progress in understanding the M-band function and its structural organization. We explain how the M-band might reduce the intrinsic instability of thick filaments and help titin to maintain order in the sarcomeres. The M-band molecular structure has been clarified recently by biochemical and biophysical approaches that focused on the properties of the prominent M-band component myomesin. These have shown that antiparallel myomesin dimers might link the thick filaments in the M-band, a role analogous to that of alpha-actinin in the Z-disc. Furthermore, similar to titin, myomesin is a molecular spring with complex visco-elastic properties that can be modified by alternative splicing. M-band protein composition correlates with the expression of titin isoforms and appears to be a reliable marker for biomechanical conditions in contracting muscle. We propose that the M-band is in fact a dynamic structure that monitors the stress appearing in the thick filament lattice during contraction and quickly reorganizes to meet new physiological requirements.     

Z band dynamics as a function of sarcomere length and the contractile state of muscle

The Z band in skeletal muscle has two distinct structural states--a relaxed (small square or ss) form and a maximally activated (basket weave or bw) form. We have examined by electron microscopy and optical diffraction Z lattice forms and dimensions and A band spacings in relaxed, tetanized, stretched, and stretched-and-tetanized rat soleus muscle. We have tested the independent contributions of passive load, active tension, and sarcomere length to Z band state. As the A band spacing decreased with increasing load and increasing sarcomere length in the untetanized muscles, the Z lattice remained in the ss form and the Z spacing changed only slightly. Computer-enhanced images from digitized electron micrographs showed that the ss Z lattice resisted deformation regardless of load or method of stretching. In contrast, when the muscle was tetanized at sarcomere lengths of up to 2.7 microns, the Z lattice assumed the bw form and the Z spacing was increased by 20%. Regardless of lattice form, Z spacing did not vary significantly with sarcomere length. Images from freeze-substituted preparations showed both lattice forms comparable to those in images from glutaraldehyde-fixed muscles. Thus, Z band state appears to be a function of the presence (or absence) of active tension. Our previous three-dimensional model is compatible with these observations and with the sub-structures revealed by computer-enhanced images of both lattice forms.     

Purification and initial characterization of a protein from skeletal muscle that caps the barbed ends of actin filaments

We describe herein the purification of a protein from skeletal muscle that binds to ("caps") the morphologically defined barbed end of actin filaments. This actin-capping protein appeared to be a heterodimer with chemically and immunologically distinct subunits of Mr = 36,000 (alpha) and 32,000 (beta), Rs = 37 A, s20,w = 4.0 S, and a calculated native molecular weight of approximately 61,000. The protein was obtained in milligram quantities at greater than 95% purity from acetone powder of chicken skeletal muscle by extraction in 0.6 M KI, precipitation with ammonium sulfate, sequential chromatographic steps on DEAE-cellulose, hydroxylapatite, and Sephacryl S-200, followed by preparative rate zonal sucrose density gradient centrifugation. In immunoblots of myofibrillar proteins, affinity-purified antibodies selectively recognized protein bands of the same molecular weight as the subunits of the capping protein to which they were made, indicating that the isolated capping protein is a native myofibrillar protein, and not a proteolytic digestion product of a larger muscle protein. A specific interaction of the capping protein with the barbed end of actin filaments was indicated by its ability to inhibit actin filament assembly nucleated by spectrin-band 4.1-actin complex in 0.4 mM Mg2+, accelerate actin filament formation and increase the critical concentration of actin in 2-5 mM Mg2+, 75-100 mM KCl, and inhibit the addition of actin monomers to the barbed end of heavy meromyosin-decorated actin filaments as determined by electron microscopy. All of these effects occurred at nanomolar concentrations of capping protein and micromolar concentrations of actin, suggesting a high affinity interaction.     

Three-dimensional structure of the Z band in a normal mammalian skeletal muscle

The three-dimensional structure of the vertebrate skeletal muscle Z band reflects its function as the muscle component essential for tension transmission between successive sarcomeres. We have investigated this structure as well as that of the nearby I band in a normal, unstimulated mammalian skeletal muscle by tomographic three- dimensional reconstruction from electron micrograph tilt series of sectioned tissue. The three-dimensional Z band structure consists of interdigitating axial filaments from opposite sarcomeres connected every 18 +/- 12 nm (mean +/- SD) to one to four cross-connecting Z- filaments are observed to meet the axial filaments in a fourfold symmetric arrangement. The substantial variation in the spacing between cross-connecting Z-filament to axial filament connection points suggests that the structure of the Z band is not determined solely by the arrangement of alpha-actinin to actin-binding sites along the axial filament. The cross-connecting filaments bind to or form a "relaxed interconnecting body" halfway between the axial filaments. This filamentous body is parallel to the Z band axial filaments and is observed to play an essential role in generating the small square lattice pattern seen in electron micrographs of unstimulated muscle cross sections. This structure is absent in cross section of the Z band from muscles fixed in rigor or in tetanus, suggesting that the Z band lattice must undergo dynamic rearrangement concomitant with crossbridge binding in the A band.     

Decellularization of porcine skeletal muscle extracellular matrix for the formulation of a matrix hydrogel: a preliminary study

Extracellular matrix (ECM) hydrogels are used as scaffolds to facilitate the repair and reconstruction of tissues. This study aimed to optimize the decellularization process of porcine skeletal muscle ECM and to formulate a matrix hydrogel scaffold. Five multi‐step methods (methods A–E) were used to generate acellular ECM from porcine skeletal muscle [rinsing in SDS, trypsin, ethylenediaminetetraacetic acid (EDTA), Triton X‐100 and/or sodium deoxycholate at 4–37°C]. The resulting ECM was evaluated using haematoxylin and eosin, 4‐6‐diamidino‐2‐phenylindole (DAPI) staining, and DNA quantification. Acellular matrix was dissolved in pepsin and gelled at 37°C. Hydrogel response to temperature was observed in vivo and in vitro. ECM components were assessed by Masson, Sirius red, and alcian blue staining, and total protein content. Acellular porcine skeletal muscle exhibited a uniform translucent white appearance. No intact nuclear residue was detected by haematoxylin and eosin staining, while DAPI staining showed a few nuclei in the matrixes produced by methods B, C, and D. Method A generated a gel that was too thin for gelation. However, the matrix obtained by rinsing in 0.2% trypsin/0.1% EDTA, 0.5% Triton X‐100, and 1% Triton X‐100/0.2% sodium deoxycholate was nuclei‐free and produced a viscous solution that formed a structurally stable white jelly‐like hydrogel. The residual DNA content of this solution was 49.37 ± 0.72 ng/mg, significantly less than in fresh skeletal muscle, and decreased to 19.22 ± 0.85 ng/mg after gelation (P < 0.05). The acellular matrix was rich in collagen and glycosaminoglycan, with a total protein concentration of 64.8 ± 6.9%. An acellular ECM hydrogel from porcine skeletal muscle was efficiently produced.     

Nebulin as a length regulator of thin filaments of vertebrate skeletal muscles: correlation of thin filament length, nebulin size, and epitope profile

Nebulin, a family of giant proteins with size-variants from 600 to 900 kD in various skeletal muscles, have been proposed to constitute a set of inextensible filaments anchored at the Z line (Wang, K., and J. Wright. 1988. J. Cell Biol. 107:2199-2212). This newly discovered filament of the skeletal muscle sarcomere is an attractive candidate for a length-regulating template of thin filaments. To evaluate this hypothesis, we address the question of coextensiveness of nebulin and the thin filament by searching for a correlation between the size of nebulin variants and the length distribution of the thin filaments in several skeletal muscles. A positive linear correlation indeed exists for a group of six skeletal muscles that display narrow thin filament length distributions. To examine the molecular and architectural differences of nebulin size-variants, we carried out immunoelectron microscopic studies to map out epitope profiles of nebulin variants in these muscles. For this purpose, a panel of mAbs to distinct nebulin epitopes was produced against rabbit nebulin purified by an improved protocol. Epitope profiles of nebulin variants in three skeletal muscles revealed that (a) nebulin is inextensible since nebulin epitopes maintain a fixed distance to the Z line irrespective of the degree of sarcomere stretch; (b) a single nebulin polypeptide spans a minimal distance of 0.9 microns from the Z line; (c) nebulin contains repeating epitopes that are spaced at 40 nm or its multiples; (d) nebulin repeats coincide with thin filament periodicity; (e) nebulin variants differ mainly at either or both ends; and (f) nebulin remains in the sarcomere in actin-free sarcomeres produced by gelsolin treatment. Together, these data suggest that nebulin is an inextensible full-length molecular filament that is coextensive with thin filaments in skeletal muscles. We propose that nebulin acts as a length-regulating template that determines thin filament length by matching its large number of 40-nm repeating domains with an equal number of helical repeats of the actin filaments.     

Interaction of alpha-actinin and nebulin in vitro. Support for the existence of a fourth filament system in skeletal muscle

Nebulin is a high molecular weight polypeptide (mass 0.6-0.8 million) which accounts for 3% of the myofibrillar mass in skeletal muscle. Due to its resistance to extraction under native conditions, relatively little is known about the biochemistry of the molecule. Here we report in vitro binding of alpha-actinin (a major Z-line protein) to nebulin. After solubilization with sodium dodecylsulfate myofibrillar polypeptides separated by gel electrophoresis were blotted on nitrocellulose and probed with 125I-labelled alpha-actinin. Nebulin is the only polypeptide decorated by alpha-actinin. This result gives biochemical support for the hypothesis, based on recent immunoelectron micrographs, that nebulin could form in skeletal muscle a fourth filament system, possibly extending to the Z-line.     

Viscoelasticity of the sarcomere matrix of skeletal muscles. The titin-myosin composite filament is a dual-stage molecular spring.

The mechanical roles of sarcomere-associated cytoskeletal lattices were investigated by studying the resting tension-sarcomere length curves of mechanically skinned rabbit psoas muscle fibers over a wide range of sarcomere strain. Correlative immunoelectron microscopy of the elastic titin filaments of the endosarcomeric lattice revealed biphasic extensibility behaviors and provided a structural interpretation of the multiphasic tension-length curves. We propose that the reversible change of contour length of the extensible segment of titin between the Z line and the end of thick filaments underlies the exponential rise of resting tension. At and beyond an elastic limit near 3.8 microns, a portion of the anchored titin segment that adheres to thick filaments is released from the distal ends of thick filament. This increase in extensible length of titin results in a net length increase in the unstrained extensible segment, thereby lowering the stiffness of the fiber, lengthening the slack sarcomere length, and shifting the yield point in postyield sarcomeres. Thus, the titin-myosin composite filament behaves as a dual-stage molecular spring, consisting of an elastic connector segment for normal response and a longer latent segment that is recruited at and beyond the elastic limit of the sarcomere. Exosarcomeric intermediate filaments contribute to resting tension only above 4.5 microns. We conclude that the interlinked endo- and exosarcomeric lattices are both viscoelastic force-bearing elements. These distinct cytoskeletal lattices appear to operate over two ranges of sarcomere strains and collectively enable myofibrils to respond viscoelastically over a broad range of sarcomere and fiber lengths.     

Novel evidence that testosterone promotes cell proliferation and differentiation via G protein-coupled receptors in the rat L6 skeletal muscle myoblast cell line

Androgens are known to modulate the skeletal muscle proliferation and differentiation processes. Recent in vitro studies have shown that dihydrotestosterone and anabolic steroids have functions in promoting the proliferation and differentiation of the mouse skeletal muscle myoblast C2C12 cell line through the classical androgen receptor (AR) signaling pathway. But there are contradictory reports that androgen plays its roles through the membrane signaling pathways. In the present study, we show that there is no expression of the classical AR in L6 cells both at gene and protein levels. We then investigated the effects of testosterone (T) on L6 cell proliferation and differentiation. The results show that T promotes L6 cell proliferation after a 24 h treatment, which followed by enhancing L6 cell differentiation, but these effects are not inhibited by flutamide (F), an antagonist of intracellular AR. Further, we tested the effect of testosterone covalently bounding to albumin (T-BSA), which does not cross the plasma membrane. The results demonstrate that T-BSA and free T have similar effects on L6 cell proliferation and differentiation, and that these effects involve G protein-coupled receptors and different downstream pathways. The L6 cell proliferation induced by T involves PKC and ERK1/2 signaling pathways and cell differentiation happens via the PKA signaling pathway. These results suggest that T promotes cell proliferation and differentiation via G protein-coupled receptors and different downstream pathways in the L6 cell line, although the related molecular mechanisms need to be elucidated in future studies.     

Assembly of avian skeletal muscle myosins: evidence that homodimers of the heavy chain subunit are the thermodynamically stable form

Using a double antibody sandwich ELISA we examined the heavy chain isoform composition of myosin molecules isolated from chicken pectoralis major muscle during different stages of development. At 2- and 40-d posthatch, when multiple myosin heavy chain isoforms are being synthesized, we detected no heterodimeric myosins, suggesting that myosins are homodimers of the heavy chain subunit. Chymotryptic rod fragments of embryonic, neonatal, and adult myosins were prepared and equimolar mixtures of embryonic and neonatal rods and neonatal and adult rods were denatured in 8 M guanidine. The guanidine denatured myosin heavy chain fragments were either dialyzed or diluted into renaturation buffer and reformed dimers which were electrophoretically indistinguishable from native rods. Analysis of these renatured rods using double antibody sandwich ELISA showed them to be predominantly homodimers of each of the isoforms. Although hybrids between the different heavy chain fragments were not detected, exchange was possible under these conditions since mixture of biotinylated neonatal rods and fluoresceinated neonatal rods formed a heterodimeric biotinylated-fluoresceinated species upon renaturation. Therefore, we propose that homodimers are the thermodynamically stable form of skeletal muscle myosin isoforms and that there is no need to invoke compartmentalization or other cellular regulatory processes to explain the lack of heavy chain heterodimers in vivo.     

Biochemical and immunohistochemical evidence for a non-muscle myosin at the neuromuscular junction in bovine skeletal muscle.

We identified 220-kD protein in bovine skeletal muscle homogenate by affinity chromatography on an agarose column and subsequent SDS-PAGE. Peptide mass fingerprinting (MALDI mass spectrometry) and internal sequence analysis revealed that this protein has homology with several members of the myosin superfamily, particularly with human cardiac beta-myosin heavy chain (beta-MHC). A rabbit polyclonal antibody against the 220-kD protein specifically stained a 220-kD band in Western blots of skeletal muscle homogenate. Immunohistochemical experiments on cryostat sections demonstrated that in skeletal muscle this protein is exclusively localized at the neuromuscular junctions, no immunoreactivity being present at the myofibril level. Because of its relative homology with cardiac beta-MHC, we also investigated the distribution of the 220-kD protein in bovine heart. In cardiac fibers, 220-kD protein-related immunoreactivity was restricted to the intercalated disks, whereas myofibrils were completely devoid of specific immunoreactivity. This distribution pattern was completely different from that of cardiac beta-MHC, which involved myofibrils. Because of the above biochemical and immunohistochemical features, the 220-kD protein we have identified is suggested to be a novel member of the non-muscle (non-sarcomeric) myosin family.     

Sarcolemmal organization in skeletal muscle lacking desmin: evidence for cytokeratins associated with the membrane skeleton at costameres

The sarcolemma of fast-twitch muscle is organized into "costameres," structures that are oriented transversely, over the Z and M lines of nearby myofibrils, and longitudinally, to form a rectilinear lattice. Here we examine the role of desmin, the major intermediate filament protein of muscle in organizing costameres. In control mouse muscle, desmin is enriched at the sarcolemmal domains that lie over nearby Z lines and that also contain beta-spectrin. In tibialis anterior muscle from mice lacking desmin due to homologous recombination, most costameres are lost. In myofibers from desmin -/- quadriceps, by contrast, most costameric structures are stable. Alternatively, Z line domains may be lost, whereas domains oriented longitudinally or lying over M lines are retained. Experiments with pan-specific antibodies to intermediate filament proteins and to cytokeratins suggest that control and desmin -/- muscles express similar levels of cytokeratins. Cytokeratins concentrate at the sarcolemma at all three domains of costameres when the latter are retained in desmin -/- muscle and redistribute with beta-spectrin at the sarcolemma when costameres are lost. Our results suggest that desmin associates with and selectively stabilizes the Z line domains of costameres, but that cytokeratins associate with all three domains of costameres, even in the absence of desmin.