Externalization of an endogenous chicken muscle lectin with in vivo development.

Chicken-lactose-lectin-I (CLL-I), an endogenous lectin that is developmentally regulated in embryonic muscle, was localized by immunohistochemical techniques in tissue samples taken at various stages of in vivo development and in primary muscle cultures. Lectin, which was diffusely distributed in myoblasts, became localized in myotubes in a distribution similar to that of the sarcoplasmic reticulum and T tubules. Later in development, lectin was predominantly extracellular. This sequence suggests that externalization may have occurred by migration in the T tubules, which are continuous with the extracellular space, although alternative explanations are possible. Only traces of lectin were found in the adult. These studies did not reveal the function of CLL-I in muscle development. However, we infer that it acts by organizing complementary glycoconjugates in the intracellular tubular network, on the muscle surface, and/or in extracellular materials.

The alteration of myosin isoform compartmentation in specific mutants of Caenorhabditis elegans.

Myosin isoforms A and B are located at the surface of the central and polar regions, respectively, of thick filaments in body muscle cells of Caenorhabditis elegans, whereas paramyosin and a distinct core structure comprise the backbones of these filaments. Thick filaments and related structures were isolated from nematode mutants that have altered thick filament protein compositions. These mutant filaments and their complexes with specific antibodies were studied by electron microscopy to determine the distribution of the two myosins. The compartmentation of the two myosin isoforms in body wall muscle thick filaments depends not only upon the intrinsic properties of the myosins but their interactions with other components such as paramyosin and their relative quantities determined by synthesis.

Rabbit skeletal muscle glycogen. A morphological and biochemical study of glycogen beta-particles isolated by the precipitation-centrifugation method.

Glycogen in its particulate beta-form is localized in the sarcoplasm close to the sarcoplasmic reticulum. Some particles are in close contact with the membranes, on the outer side of the vesicles. The mild technique of differential precipitation-centrifugation has been adapted to the preparation of glycogen from adult skeletal muscle. A preliminary low-speed centrifugation which eliminates the contractile protein structures and the cell debris is followed by a high-speed centrifugation which produces pellets containing glycogen mixed with smooth-walled vesicles, the glycogen-sarcovesicular fraction. The glycogen obtained after treatment of this fraction with deoxycholate and two washings contains 3% protein. A similar protein content contaminates glycogen banded in a linear sucrose gradient. The glycogen-sarcovesicular fraction and the purified glycogen have been examined, under the electron microscope, in sections of fixed and embedded material or with the negative staining technique. The glycogen beta-particles in negatively stained preparations have an average diameter of 39.4 mmicro. The largest particles present irregular outlines, suggesting the presence of conglomerated subunits, about 20 mmicro in diameter. These subunits seem to fall apart under the influence of concentrated potassium hydroxide. The mean sedimentation coefficients calculated for infinite dilution vary from 115 to 135S. The spectrophotometric analysis of the glycogen-iodine complex indicates the presence of long end-chains in the molecule.

Multiple actins in Drosophila melanogaster.

The tissue and developmental specificities of the three Drosophila isoactins, originally identified in primary myogenic cultures and in the permanent Schneider L-2 cell line, have been investigated. Of these three isoactins (I, II, and III), actins I and II are stable and actin III is unstable. Two-dimensional polyacrylamide gel electrophoretic analyses of total cellular extracts after 1-h [(35)S]methionine pulses were performed on a large variety of embryonic, larval, and adult muscle and nonmuscle tissues. The results suggest that isoactins II and III are generalized cellular actins found in all drosophila cell types. Actin I, on the other hand, is muscle-associated and is found exclusively in supercontractile muscle (such as larval body wall and larval and adult viscera) including primary myogenic cell cultures. Although actin I synthesis is not detectable during very early embryogenesis, it is detectable by 25 h and actin I is a major stable actin in all larval muscle tissues. Actin I is synthesized in reduced amounts relative to the other actins in late third instar larvae but is again a major product of actin synthesis in the adult abdomen. A stable actin species with the same pI as actin III has been identified in the adult thorax and appears to be unique to flight muscle tissue. This new stable form of thoracic actin may be the result of a stabilization of the actin III found in other tissues or may be an entirely separate gene product.

Plasminogen activator in mammalian skeletal muscle: characteristics of effect of denervation on urokinase-like and tissue activator.

Analyses were made of the fibrinolytic, plasminogen-activating system in skeletal muscle to determine if a regulating influence of the nerve could be detected on these enzymes. Young male mice underwent right sciatic neurectomy. Extracts were prepared from denervated muscle at 2-17 d after axotomy and compared with controls. Using a cascade-style biochemical assay (Rånby, M., B. Norrman, and P. Wallén, 1982, Thromb. Res., 27:743-748) we found that low levels of plasminogen activator (PA) were present in adult, innervated mouse muscle, but that denervation resulted in a marked time-dependent increase in enzyme activity. Qualitative separation showed an eightfold increase in urokinase-like PA with moderate elevation of tissue PA activity after 10 d. Fibrin zymography (Granelli-Piperno, A., and E. Reich, 1978, J. Exp. Med., 148:223-234) revealed clear zones of lysis corresponding to molecular masses of 48 kD for urokinase-like PA and 75 kD for tissue PA, consistent with the molecular masses found for these enzymes in other tissues of the mouse (Danø, K., P. A. Andreasen, J. Grøndahl-Hansen, P. Kristensen, L. S. Nielsen, and L. Skriver, 1985, Adv. Cancer Res., 44:139-266). In other studies we have shown that PA-activated plasmin readily attacks critical adhesive basement membrane molecules. The present results indicate that enzymes involved in plasminogen activation, particularly urokinase-like PA, rapidly increase after axotomy, suggesting they may have a role early in muscle denervation. Similar alterations in PA activity might underlie the elimination of polyneuronal innervation during mammalian muscle development. Certain neuromuscular diseases may also involve activation of these enzymes, resulting in degradation of basement membrane zone components and, therefore, warrant further study.

Isolation, characterization, and localization of the spanning protein from skeletal muscle triads.

A monoclonal antibody has been developed against the putative junctional protein or spanning protein (SP) from skeletal muscle triads. By immuno-affinity chromatography, we have purified this protein. The native protein has a molecular mass of 630-800 kD, as determined by gel filtration and rate zonal centrifugation. Within the limits of the methods used, the basic unit of the SP appears to be a dimer. In electron micrographs, it is shown to exhibit a circular profile with a diameter of approximately 100 A. In thin section analysis, the protein is frequently observed as parallel tracks of electron-dense particles bordering a translucent core. We suggest that the basic unit of the junctional structure is a dimer of 300-kD subunits and that four such entities constitute the intact SP. The purified protein has been used to develop polyclonal antibodies. By immunoelectron microscopy using immunogold probes, the SP has been localized to the junctional gap of the triad. By attaching the SP to an affinity resin, three proteins have been identified as forming associations with the SP. The Mrs of the proteins are 150, 62, and 38 kD; the 62-kD protein is calsequestrin.

Stereological analysis of mammalian skeletal muscle. I. Soleus muscle of the adult guinea pig.

A quantitative analysis of the volumes, surface areas, and dimensions of the ultrastructural components in the soleus muscle fibers of the guinea pig was made by using point counting methods of stereology. Muscle fibers have structural orientation (anisotropy) and have spatial gradients of the structures within the fiber; therefore the standard stereological methods were modified where necessary. The entire analysis was repeated at two section orientations to test the modifications and identical results obtained from both. The volume of lipid droplets was 0.20 +/- 0.06% (mean +/- standard error, n = 5 animals) and the nuclei volume was 0.86 +/- 0.20% of the fiber volume. The total mitochondrial volume was 4.85 +/- 0.66% of the fiber volume with about one-third being found in an annulus within 1 microm of the sarcolemma. The mitochondrial volume in the remaining core of the fiber was 3.6 +/- 0.4%. The T system has a volume of 0.14 +/- 0.01% and a surface area of 0.064 +/- 0.005 microm(2)/microm(3) of the fiber volume. The surface area of the sarcolemma is 0.116 +/- 0.013 microm(2)/microm(3) which is twice the T system surface area. The volume of the entire sarcoplasmic reticulum is 3.52 +/- 0.33% and the surface area is 0.97 +/- 0.09 microm(2)/microm(3). The sarcoplasmic reticulum is composed of the terminal cisternae whose volume is 1.04 +/- 0.19% and surface area is 0.24 +/- 0.05 microm(2)/microm(3). The tubules of the sarcoplasmic reticulum in the I band and A band have volumes of 1.97 +/- 0.24% and 0.51 +/- 0.08%, and the surface areas of the I and A band reticulum are 0.56 +/- 0.07 microm(2)/microm(3) and 0.16 +/- 0.04 microm(2)/microm(3), respectively. The Z line width, myofibril and fiber diameters were measured.

Glutamate uptake by a stimulated insect nerve muscle preparation.

Recent reports suggest that glutamate may be the excitatory neuromuscular transmitter in insects. In this study, glutamate uptake by isolated cockroach nerve muscle preparations was investigated by means of chemical and electron microscope radioautographic techniques. We found that the preparation had a high affinity for glutamate and that nerve stimulation enhanced glutamate uptake. Chemical studies showed that the average tissue concentration of glutamate bound during a 1 hr incubation period in 10(-5)M glutamate-(3)H after nerve stimulation was 2.8 x 10(-5)M. Less than 1% of the radioactivity was present in the perchloric acid-precipitated protein fraction. Using electron microscope radioautography, we observed that sheath cells showed the highest glutamate concentration of all cellular compartments. Uptake was greater at neuromuscular junctions than in other regions of the tissue. The data suggest a possible mechanism for transmitter inactivation and protection of synapses from high blood glutamate.

Expression of cell-adhesion molecules in embryonic induction. I. Morphogenesis of nestling feathers.

The potential relationship of cell adhesion to embryonic induction during feather formation was examined by immunohistochemical analysis of the spatiotemporal distribution of three cell-adhesion molecules (CAMs), neural CAM (N-CAM), liver CAM (L-CAM), and neuron-glia CAM (Ng-CAM), and of substrate molecules (laminin and fibronectin) in embryonic chicken skin. The N-CAM found at sites of embryonic induction in the feather was found to be similar to brain N-CAM as judged by immuno-cross-reactivity, migratory position in PAGE, and the presence of embryonic to adult conversion. In contrast to the N-CAM found in the brain, however, only one polypeptide of Mr 140,000 was seen. N-CAM-positive dermal condensations were distributed periodically under L-CAM-positive feather placodes at those sites where basement membranes are known to be disrupted. After initiation of induction, L-CAM-positive placode cells became transiently N-CAM-positive. N-CAM was asymmetrically concentrated in the dorsal region of the feather bud, while fibronectin was concentrated in the ventral region. During feather follicle formation, N-CAM was expressed in the dermal papilla and was closely apposed to the L-CAM-positive papillar ectoderm, while the dermal papilla showed no evidence of laminin or fibronectin. The collar epithelium was both N-CAM- and L-CAM-positive. During the formation of the feather filament, N-CAM appeared periodically and asymmetrically on basilar cells located in the valleys between adjacent barb ridges. In contrast to the two primary CAMs, Ng-CAM was found only on nerves supplying the feather and the skin. These studies indicate that at each site of induction during feather morphogenesis, a general pattern is repeated in which an epithelial structure linked by L-CAM is confronted with periodically propagating condensations of cells linked by N-CAM.

Improved procedures for immunoferritin labeling of ultrathin frozen sections.

In employing fixed frozen ultrathin sections as substrates for immunoferritin labeling of intracellular antigens, we have found that conventional glutaraldehyde fixation sometimes permits very little specific staining of the sections, either because it inactivates certain protein antigens, or because it renders them inaccessible to the antibody stains. We have developed several fixation procedures that are chemically milder and allow a uniform but less extensive cross-linking of the specimen. With these procedures and precautions in the handling of the more fragile frozen sections, excellent structural preservation and specific immunoferritin labeling has been achieved with several systems.

Development of muscle fiber specialization in the rat hindlimb.

The appearance of fast and slow fiber types in the distal hindlimb of the rat was investigated using affinity-purified antibodies specific to adult fast and slow myosins, two-dimensional electrophoresis of myosin light chains, and electron microscope examination of developing muscle cells. As others have noted, muscle histogenesis is not synchronous; rather, a series of muscle fiber generations occurs, each generation forming along the walls of the previous generation. At the onset of myotube formation on the 15th d of gestation, the antimyosin antibodies do not distinguish among fibers. All fibers react strongly with antibody to fast myosin but not with antibody to slow myosin. The initiation of fiber type differentiation can be detected in the 17-d fetus by a gradual increase in the binding of antibody to slow myosin in the primary, but not the secondary, generation myotubes. Moreover, neuromuscular contacts at this crucial time are infrequent, primitive, and restricted predominantly, but not exclusively, to the primary generation cells, the same cells which begin to bind large amounts of antislow myosin at this time. With maturation, the primary generation cells decrease their binding of antifast myosin and become type I fibers. Secondary generation cells are initially all primitive type II fibers. In future fast muscles the secondary generation cells remain type II, while in future slow muscles most of the secondary generation cells eventually change to type I over a prolonged postnatal period. We conclude that the temporal sequence of muscle development is fundamentally important in determining the genetic expression of individual muscle cells.

Distribution of fast myosin heavy chain isoforms in thick filaments of developing chicken pectoral muscle.

Colloidal gold-conjugated monoclonal antibodies were prepared to stage-specific fast myosin heavy chain (MHC) isoforms of developing chicken pectoralis major (PM). Native thick filaments from different stages of development were reacted with these antibodies and examined in the electron microscope to determine their myosin isoform composition. Filaments prepared from 12-d embryo, 10-d chick, and 1-yr chicken muscle specifically reacted with the embryonic (EB165), neonatal (2E9), and adult (AB8) antimyosin gold-conjugated monoclonal antibodies, respectively. The myosin isoform composition was more complex in thick filaments from stages of pectoral muscle where more than one isoform was simultaneously expressed. In 19-d embryo muscle where both embryonic and neonatal isoforms were present, three classes of filaments were found. One class of filaments reacted only with the embryonic antibody, a second class reacted only with the neonatal-specific antibody, and a third class of filaments were decorated by both antibodies. Similar results were obtained with filaments prepared from 44-d chicken PM where the neonatal and adult fast MHCs were expressed. These observations demonstrate that two myosin isoforms can exist in an individual thick filament in vivo. Immunoelectron microscopy was also used to determine the specific distribution of different fast MHC isoforms within individual filaments from different stages of development. The anti-embryonic and anti-adult antibodies uniformly decorated both homogeneous and heterogeneous thick filaments. The neonatal specific antibody uniformly decorated homogeneous filaments; however, it preferentially decorated the center of heterogeneous filaments. These observations suggest that neonatal MHC may play a specific role in fibrillogenesis.

A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies.

The M line, which transverses the center of the thick filament region of skeletal muscle sarcomeres, appears to be a complex array of multiple structural elements. To date, two proteins have definitely been shown to be associated with the M line. They are MM-CK, localized in the M 4,4' substriations, and a 165,000-dalton (164 kd) protein, referred to as both M-protein and myomesin. Here we report the positive identification of a third M-line protein of 185 kd. In the course of making monoclonal antibodies (mAbs) against a 165-kd fraction, we also obtained mAbs that bound to the M line of isolated myofibrils as detected by indirect immunofluorescence, but recognized a protein band of 185 kd in immunoblotting experiments with either the original immunogen or low ionic strength myofibril extracts as antigenic targets. The evidence that the 185- and 165-kd proteins are distinct protein species is based on the separation of the two proteins into discrete peaks by ion exchange chromatography, the distinctive patterns of their degradation products, and non-cross-reactivity of any of seven mAbs. These mAbs recognize three unique antigenic determinants on the 185-kd molecule and at least two and probably four sites on the 165-kd molecule as determined from competitive binding and immunofluorescence experiments. To resolve the problem of multiple nomenclature for the 165-kd protein, the 185-kd protein will be referred to as myomesin and the 165-kd protein as M-protein.

Polymorphism of myosin among skeletal muscle fiber types.

An immunocytochemical approach was used to localize myosin with respect to individual fibers in rat skeletal muscle. Transverse cryostat sections of rat diaphragm, a fast-twitch muscle, were exposed to fluorescein-labeled immunoglobulin against purified chicken pectoralis myosin. Fluorescence microscopy revealed a differential response among fiber types, identified on the basis of mitochondrial content. All white and intermediate fiber but only about half of the red fiber reacted with his antimyosin. In addition, an alkali-stable ATPase had the same pattern of distribution among fibers, which is consistent with the existence of two categories of red fibers. The positive response of certain red fibers indicates either that their myosin has antigenic determinants in common with "white" myosin, or that the immunogen contained a "red" myosin. Myosin, extracted from a small region of the pectorlis which consists entirely of white fibers, was used to prepare an immunoadsorbent column to isolate antibodies specific for white myosin. This purified anti-white myosin reacted with the same fibers of the rat diaphragm that had reacted with the white, intermediate, and some red fibers are sufficiently homologous to share antigenic determinants. In a slow-twitch muscle, the soleus, only a minority of the fiber reacted with antipectoralis myosin. The majority failed to respond; hence, they are not equivalent to intermediate fibers of the diaphragm; despite their intermediate mitochondrial content. Immunocytochemical analysis of two different musles of the rat has demonstrated that more than one isoenzyme of myosin can exist in a single muscle, and that individual fiber types can be recognized by immunological differences in their myosin. We conclude that, in the rat diaphragm, there are at least two immunochemically distinct types of myosin and four types of muscle fibers: white, intermediate, and two red. We suggest that these fibers correspond to the four types of motor units described by Burke et al. (Burke, R. E., D. N. Levine, P. Tsairis, and F. E. Zajac, III 1973. J. Physiol. (Lond) 234:723-748.)in the cat gastrocnemius.;

Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro.

Monoclonal antibodies (McAbs) against the myosin heavy chain (MHC) of adult chicken pectoralis muscle have been tested for reactivity with pectoralis myosin at selected stages of chick development in vivo and in vitro. Three such McAbs, MF 20 and MF 14, which bind to light meromyosin, and MF 30, which binds to myosin subfragment two (S2), were used to assay the appearance and accumulation of specific MHC epitopes with: (a) indirect, solid phase radioimmune assay (RIA), (b) immunoautoradiography, (c) immunofluorescence microscopy. McAb MF 20 bound strongly and equivalently to MHC at all stages of embryonic development in vivo. In contrast, the MF 30 epitope was barely detectable at 12 d of incubation but its concentration rose rapidly just before hatching. No detectable binding of MF 14 to pectoralis myosin could be measured during myogenesis in vivo until 1 wk after hatching. Immunofluorescence studies revealed that all three epitopes accumulate in the same myocytes of the developing pectoralis muscle. Since all three McAbs bound with high activity to native and denatured forms of myosin, it is unlikely that differential antibody reactivity can be explained by conformational changes in myosin during development in vivo. When myogenesis in vitro was monitored using the same McAbs, MF 20 bound to the MHC at all stages tested while reactivity of MF 30 and MF 14 with myosin from cultured muscle was never observed. Thus, this study demonstrates three different immunochemical states of the MHC during development in vivo of chick pectoralis muscle and the absence of later occurring immunochemical transitions in the MHC of cultured embryonic muscle.

Brush-border alpha-actinin? Comparison of two proteins of the microvillus core with alpha-actinin by two-dimensional peptide mapping.

The bundle of filaments within the intestinal microvillus contains four major polypeptides in addition to actin calmodulin, a 70-kdalton subunit and two polypeptides with molecular masses similar to that of the Z-line component alpha-actinin (95 and 105 kdaltons). Two-dimensional mapping of tryptic peptides indicates that (a) alpha-actinins from chicken skeletal, cardiac, and smooth muscle are similar but not identical proteins and that skeletal alpha-actinin in more similar to the cardiac subunit than to the alpha-actinin from gizzard; (b) the brush-border 95- and 105-kdalton subunits are closely related to each other, but the smaller subunit is not a proteolytic fragment of the 105-kdalton subunit; and (c) although there is considerable peptide overlap between the brush-border subunits and the three alpha-actinins, the peptide maps of the 95- and 105-kdalton proteins are substantially distinct from the various alpha-actinin maps, suggesting that neither brush-border subunit is a bona fide alpha-actinin. Nevertheless, on the basis of peptide mapping criteria alone, one cannot exclude the possibility that the brush-border subunits are "alpha-actinin-like." However, there is no immunological cross-reactivity between the brush-border subunits and alpha-actinins, using antibodies prepared against gizzard alpha actinin.

Monoclonal antibodies distinguish titins from heart and skeletal muscle.

Murine monoclonal antibodies specific for titin have been elicited using a chicken heart muscle residue as antigen. The three antibodies T1, T3, and T4 recognize both bands of the titin doublet in immunoblot analysis on polypeptides from chicken breast muscle. In contrast, on chicken cardiac myofibrils two of the antibodies (T1, T4) react only with the upper band of the doublet indicating immunological differences between heart and skeletal muscle titin. This difference is even more pronounced for rat and mouse. Although all three antibodies react with skeletal muscle titin, T1 and T4 did not detect heart titin, whereas T3 reacts with this titin both in immunofluorescence microscopy and in immunoblots. Immunofluorescence microscopy of myofibrils and frozen tissues from a variety of vertebrates extends these results and shows that the three antibodies recognize different epitopes. All three titin antibodies decorate at the A-I junction of the myofibrils freshly prepared from chicken skeletal muscle and immunoelectron microscopy using native myosin filaments demonstrates that titin is present at the ends of the thick filaments. In chicken heart, however, antibodies T1 and T4 stain within the I-band rather than at the A-I junction. The three antibodies did not react with any of the nonmuscle tissues or permanent cell lines tested and do not decorate smooth muscle. In primary cultures of embryonic chicken skeletal muscle cells titin first appears as longitudinal striations in mononucleated myoblasts and later at the myofibrillar A-I junction of the myotubes.

Cytoplasmic myosin from Drosophila melanogaster.

Myosin is identified and purified from three different established Drosophila melanogaster cell lines (Schneider's lines 2 and 3 and Kc). Purification entails lysis in a low salt, sucrose buffer that contains ATP, chromatography on DEAE-cellulose, precipitation with actin in the absence of ATP, gel filtration in a discontinuous KI-KCl buffer system, and hydroxylapatite chromatography. Yield of pure cytoplasmic myosin is 5-10%. This protein is identified as myosin by its cross-reactivity with two monoclonal antibodies against human platelet myosin, the molecular weight of its heavy chain, its two light chains, its behavior on gel filtration, its ATP-dependent affinity for actin, its characteristic ATPase activity, its molecular morphology as demonstrated by platinum shadowing, and its ability to form bipolar filaments. The molecular weight of the cytoplasmic myosin's light chains and peptide mapping and immunochemical analysis of its heavy chains demonstrate that this myosin, purified from Drosophila cell lines, is distinct from Drosophila muscle myosin. Two-dimensional thin layer maps of complete proteolytic digests of iodinated muscle and cytoplasmic myosin heavy chains demonstrate that, while the two myosins have some tryptic and alpha-chymotryptic peptides in common, most peptides migrate with unique mobility. One-dimensional peptide maps of SDS PAGE purified myosin heavy chain confirm these structural data. Polyclonal antiserum raised and reacted against Drosophila myosin isolated from cell lines cross-reacts only weakly with Drosophila muscle myosin isolated from the thoraces of adult Drosophila. Polyclonal antiserum raised against Drosophila muscle myosin behaves in a reciprocal fashion. Taken together our data suggest that the myosin purified from Drosophila cell lines is a bona fide cytoplasmic myosin and is very likely the product of a different myosin gene than the muscle myosin heavy chain gene that has been previously identified and characterized.

Intermediate filaments in muscle and epithelial cells of nematodes.

Current concepts of the developmentally controlled multigene family of intermediate filament (IF) proteins expect the origin of their complexity in evolutionary precursors preceding all vertebrate classes. Among invertebrates, however, firm ultrastructural as well as molecular documentation of IFs is restricted to some giant axons and to epithelia of a few molluscs and annelids. As Ascaris lumbricoides is easily dissected into clean tissues, IF expression in this large nematode was analyzed by electron microscopic and biochemical procedures and a monoclonal antibody reacting with all mammalian IF proteins. We document for the first time the presence of IFs in muscle cells of an invertebrate. They occur in three muscle types (irregular striated pharynx muscle, obliquely striated body muscle, uterus smooth muscle). IFs are also found in the epithelia studied (syncytial epidermis, intestine, ovary, testis). Immunoblots on muscles, pharynx, intestine, uterus, and epidermis identify a pair of polypeptides (with apparent molecular masses of 71 and 63 kD) as IF constituents. In vitro reconstitution of filaments was obtained with the proteins purified from body muscle. In the small nematode Caenorhabditis elegans IF proteins are so far found only in the massive desmosome-anchored tonofilament bundles which traverse a special epithelial cell type, the marginal cells of the pharynx. We speculate that IFs may occur in most but perhaps not all invertebrates and that they may not occur in all cells in large amounts. As electron micrographs of the epidermis of a planarian--a member of the Platyhelminthes--reveal IFs, the evolutionary origin of this cytoplasmic structure can be expected either among the lowest metazoa or already in some unicellular eukaryotes.

Agrin-like molecules at synaptic sites in normal, denervated, and damaged skeletal muscles.

Several lines of evidence have led to the hypothesis that agrin, a protein extracted from the electric organ of Torpedo, is similar to the molecules in the synaptic cleft basal lamina at the neuromuscular junction that direct the formation of acetylcholine receptor and acetylcholinesterase aggregates on regenerating myofibers. One such finding is that monoclonal antibodies against agrin stain molecules concentrated in the synaptic cleft of neuromuscular junctions in rays. In the studies described here we made additional monoclonal antibodies against agrin and used them to extend our knowledge of agrin-like molecules at the neuromuscular junction. We found that anti-agrin antibodies intensely stained the synaptic cleft of frog and chicken as well as that of rays, that denervation of frog muscle resulted in a reduction in staining at the neuromuscular junction, and that the synaptic basal lamina in frog could be stained weeks after degeneration of all cellular components of the neuromuscular junction. We also describe anti-agrin staining in nonjunctional regions of muscle. We conclude the following: (a) agrin-like molecules are likely to be common to all vertebrate neuromuscular junctions; (b) the long-term maintenance of such molecules at the junction is nerve dependent; (c) the molecules are, indeed, a component of the synaptic basal lamina; and (d) they, like the molecules that direct the formation of receptor and esterase aggregates on regenerating myofibers, remain associated with the synaptic basal lamina after muscle damage.