Expression patterns of initiator and effector caspases in denervated human skeletal muscle

There is evidence that apoptotic cell death contributes to the loss of denervated muscle fibers. In 17 patients with neurogenic muscular atrophy, we studied the expression of the apoptosis mediators APAF-1/caspase-9 and degrading caspases-2, -3, and -7 by immunohistochemical and western blot analyses. Muscle with neurogenic atrophy showed distinct upregulation of caspase-9 and -7 and no expression for APAF-1 (apoptosis protease-activating factor-1) and caspase-2 and -3. Expression of caspase-7 was restricted to atrophic fibers, but caspase-9 was also found in normal-sized muscle fibers where its expression was often confined to single fiber segments. These findings indicate that upregulated expression of caspase-9 can initiate the proteolytic cascade involving the downstream executioner caspase-7, which mediates degradation of denervated muscle fibers. However, apoptotic events may be restricted to single muscle-fiber segments, where apoptotic cell degradation contributes to the long-term process of atrophy. Pharmacological inhibition of caspases may be a therapeutic strategy in diminishing muscle atrophy.     

Localization of survival motor neuron protein in human apoptotic-like and regenerating muscle fibers, and neuromuscular junctions.

Mutations in the gene encoding survival motor neuron (SMN) protein are found in > 98% of patients with autosomal-recessive spinal muscular atrophy. We investigated the possible role of SMN in normal and abnormal human muscle by immunostaining biopsies of 20 patients with various neuromuscular diseases using monoclonal antibodies against SMN. SMN was strongly expressed cytoplasmically in chronic peripheral neuropathies, in about 80% of chronically denervated, very atrophic muscle fibers containing clumps of TUNEL-positive pyknotic nuclei: about 60% of those fibers also had cytoplasmic Bcl-2 and Bax immunoreactivity. In regenerating muscle fibers of various myopathies SMN co-localized with desmin, Bcl-2 and Bax; it was also present at the postsynaptic domain of normal human neuromuscular junctions. Thus, SMN may play a role in normal and pathological processes of adult human muscle fibers.     

Effects of passive exercise on neurogenic atrophy in rat skeletal muscle

Passive exercise treatment for 23 days produced a retardation of type II muscle fiber atrophy in denervated extensor digitorum longus muscle of rat compared with denervated-nontreated animals. The type I muscle fibers of both denervated groups were similar to that of control rats.     

Embryonic neurons transplanted into the tibial nerve reinnervate muscle and reduce atrophy but NCAM expression persists

OBJECTIVE: The aim of this study was to use the glycogen depletion technique to determine whether reinnervated muscle fibers could be distinguished from denervated muscle fibers by their size or by neural cell adhesion molecule (NCAM) expression. METHODS: Medial gastrocnemius muscles of five adult Fischer rats were reinnervated from embryonic neurons transplanted into the distal stump of the tibial nerve. Ten weeks later, the transplants were stimulated repeatedly to deplete reinnervated muscle fibers of glycogen. Areas of reinnervated (glycogen-depleted) muscle fibers were measured and assessed for NCAM expression. The areas of muscle fibers from reinnervated, denervated (n=5) and unoperated control muscles (n=5) were compared. RESULTS: Mean reinnervated muscle fiber area was significantly larger than the mean for denervated fibers (mean +/- SE: 40 +/- 6 and 10 +/- 1% of unoperated control fibers, respectively). NCAM was expressed in 55 +/- 7% of reinnervated fibers (mean +/- SE; range: 42-77%). The mean areas of reinnervated fibers that did or did not express NCAM were similar. NCAM was only expressed in some fibers in completely denervated muscles. DISCUSSION: Our data show that NCAM expression does not differentiate muscle denervation or reinnervation. Quantifying the area of large fibers did distinguish reinnervated muscle fibers from denervated fibers and showed that reinnervation of muscle from neurons placed in peripheral nerve is a strategy to rescue muscle from atrophy.     

Coexpression of myosin isoforms in muscle of patients with neurogenic disease

Three well-characterized antimyosin heavy chain monoclonal antibodies (McAbs) were used as immunocytochemical reagents to study myosin isoform expression in relationship to adenosine triphosphatase (ATPase) defined fiber types in human muscle. The biopsy specimens were from patients with neurogenic muscle disease whose muscle exhibited fiber type grouping and group atrophy. The use of McAbs revealed heretofore unrecognized coexpression of multiple myosin isoforms in selected fibers in the pathologic samples which was not apparent with ATPase reactions and not present in normal muscle. The fibers containing multiple myosin isoforms were probably undergoing neurally directed fiber type transformation. Furthermore, a small population of fibers in neurogenic specimens expressed a "prenatal" myosin signifying the presence of regenerating fibers. We also demonstrated immunocytochemical evidence of the persistence of adult slow myosin in denervated mature human skeletal muscle despite the reputed necessity of innervation for maintenance of expression of this myosin isoform proffered by others.     

Interleukin-1 expression in normal motor endplates and muscle fibers showing neurogenic changes

Histological features of neurogenic muscle involvement include type grouping, muscle fiber atrophy and target fibers. In zidovudine-induced myopathy and dermatomyositis, immunoreactivity for interleukin (IL)-1 has been reported in diseased muscle fibers involving myofibrillar breakdown and atrophy. Since IL-1 is a signal for muscle proteolysis, we studied myofiber expression of IL-1 in neurogenic muscle involvement, specially in atrophic myofibers and target fibers which are associated with myofilament breakdown. Muscle biopsy samples from patients with normal (5 cases) or neurogenic muscle involvement (25 cases) were studied by enzyme histochemistry and immunohistochemistry. In normal muscles, immunoreactivity for IL-1β was restricted to the post-synaptic domain of motor endplates and that for IL-1α had a similar localization but was faint. Immunoreactivity for IL-1α and -β was observed, respectively, in 42.5% and 75.5% of target fibers, in 8.5% and 10.4% of dark angulated fibers, in 0% and 0.3% of non-atrophic type-grouped fibers, in 14.2% and 16.5% of moderately atrophic fibers, and in 65% and 20.9% of severely atrophic fibers. Immunoblot study showed the presence of both proIL-1 (31 kDa) and mature IL-1 (17.5 kDa). From this study, we conclude that IL-1 is normally expressed in the muscular domain of neuromuscular junctions; that IL-1 is mainly expressed in neurogenic target fibers; and that IL-1 expression by muscle fibers in pathological conditions seems to be associated with myofibrillar protein breakdown and regeneration. Received: 14 October 1996 / Revised, accepted: 5 February 1997     

Electrical stimulation based on chronaxie reduces atrogin-1 and myoD gene expressions in denervated rat muscle

Denervation induces muscle fiber atrophy and changes in the gene expression rates of skeletal muscle. Electrical stimulation (ES) is a procedure generally used to treat denervated muscles in humans. This study evaluated the effect of ES based on chronaxie and rheobase on the expression of the myoD and atrogin-1 genes in denervated tibialis anterior (TA) muscle of Wistar rats. Five groups were examined: (1) denervated (D); (2) D+ES; (3) sham denervation; (4) normal (N); and (5) N+ES. Twenty muscle contractions were stimulated every 48 h using surface electrodes. After 28 days, ES significantly decreased the expression of myoD and atrogin-1 in D+ES compared to the D group. However, ES did not prevent muscle-fiber atrophy after denervation. Thus, ES based on chronaxie values and applied to denervated muscles using surface electrodes, as normally used in human rehabilitation, was able to reduce the myoD and atrogin-1 gene expressions, which are related to muscular growth and atrophy, respectively. The results of this study provide new information for the treatment of denervated skeletal muscle using surface ES.     

Effect of nerve extract on atrophy of denervated or immobilized muscles

Changes in denervated muscles are due to disuse caused by paralysis of the muscle and the loss of special neurotrophic substances. We determined the relative roles of these two factors in the production of atrophy in denervated rats' extensor digitorum longus (EDL) muscles. Muscles were denervated and/or immobilized (by fixation of the ankle) for 7 days. Some rats also received daily intramuscular injections of a saline extract of rats' sciatic nerves (2.0 mg protein/ml). Atrophy was assessed by measurement of wet weight, total protein, and cross-sectional areas of types IIA and IIB fibers (in sections stained for ATPase). Both denervation and immobilization produced significant decreases in weight, protein, and areas of fiber. The group of rats with denervated EDL muscles had significantly greater atrophy than the group with immobilized muscles. In another group, denervated EDL muscles had significantly greater atrophy than contralateral muscles which were immobilized. However, when denervated muscles were injected with nerve extract, they did not differ significantly from contralateral, noninjected, immobilized muscles. Comparisons of the group of rats in which one EDL was denervated with groups in which one muscle was immobilized or was denervated and injected with nerve extract, indicated that loss of trophic influence was responsible for about 40% of the decreases in wet weight, total protein, and cross-sectional area of type IIB fibers, and the remaining 60% was due to disuse. Loss of trophic influence was responsible for only about 5% of the atrophy of denervated type IIA fibers. Therefore, inactivity and loss of neurotrophic influence were responsible for the atrophy which occurred in denervated skeletal muscles, and these two factors influenced the two types of fiber differently. The component of denervation atrophy due to loss of trophic influence could be completely prevented by injection of substances extracted from peripheral nerves.     

Tenascin and laminin function in target recognition and central synaptic differentiation

The influence of the target cell-issued extracellular molecules tenascin-C and laminin on synaptogenesis was studied in mixed primary cultures of pituitary melanotrophs and hypothalamic neurons. We could demonstrate in this neuron-target co-culture system a new role for tenascin-C, which appeared to be expressed as an early and transitory signal of target recognition for selective afferent fibers. Tenascin-C expression disappeared from the melanotrophs soon after the establishment of neural contacts. Concomitantly, the melanotrophs became immunoreactive for laminins, and more specifically for the synaptic isoform beta2 chain-containing laminin. The laminin signal appeared to be involved in the induction of synaptic differentiation, selectively with fibers containing both dopamine and GABA, like those innervating the melanotrophs in situ.     

Fibrillation potential amplitude to quantitatively assess denervation muscle atrophy

Denervated muscle fibers exhibit spontaneous, repetitive single muscle fiber discharges and display fibrillation potentials detectable by electromyography. To explore the changing pattern of fibrillation potential amplitude after peripheral nerve injury and its relationship to the degree of muscle atrophy, fibrillation potential amplitudes were recorded on completely denervated biceps brachii of 173 patients with brachial plexus injury. Biceps brachii biopsies were taken at the same sites as the electromyogram recordings in 63 patients. The biopsies were analyzed by ATPase staining and the cross-sectional areas of fast and slow-twitch fibers were calculated. We found that the fibrillation potential amplitude and the cross-sectional areas of denervated muscle decay over time (P < 0.05), and both correlate negatively with denervation time (P < 0.01-0.05) within the first 15 months. The fibrillation potential amplitude correlates positively with both type I and II fiber cross-sectional areas (P < 0.0005-0.01). Our results show that fibrillation potential amplitude is closely correlated with muscle fiber size during the first 15 months after nerve injury, and it may therefore serve as a convenient index to evaluate quantitatively the degree of atrophy of denervated muscles. Electromyographic studies thus may help in designing treatment strategies.     

Fiber types in normal and neonatally denervated fast muscles of the rat: Immunocytochemical study with an antimyosin monoclonal antibody

The differentiation of fiber types in normal and neonatally denervated gastrocnemius muscles of the rat was compared by myosin ATPase histochemistry and immunocytochemistry using a monoclonal antibody, HM-1.2. The specificity of HM-1.2 for the fast myosin heavy chain was determined by radioimmunoassay, immunoautoradiography, and indirect immunofluorescence techniques. In normal 1-month-old and adult rats, the type IIB (fast glycolytic) fibers of the gastrocnemius could be clearly divided into three subtypes by their graded immunofluorescence staining with the myosin heavy chain-specific monoclonal antibody. In the gastrocnemius muscle of the newborn rat, all fibers were negative with the monoclonal antibody. The transition from negative to three grades of immunoreactivity occurred 1 to 2 weeks postnatally. After neonatal denervation of the gastrocnemius muscle, however, uniformly positive monoclonal antibody immunofluorescence staining for the myosin heavy chain was observed without subtype differentiation. This study, thus, gave clear immunocytochemical evidence that the type IIB muscle fibers are heterogeneous with respect to their myosin isoform and that the expression of this heterogeneity is dependent on the normal developmental influence of motor innervation on the muscle fibers.     

Influence of locomotor training on the structure and myosin heavy chains of the denervated rat soleus muscle

OBJECTIVE: Mechanism of denervation atrophy remains poorly understood. In particular, the question about irreversibility of the late atrophy is still open. Therefore, in the present study, we investigated whether and how a passive movement can affect a progress of atrophy in rat soleus muscle. To address this issue, a locomotor training on a treadmill was applied to rats with their right hindlimb muscles denervated. METHODS: The hindlimb muscles were denervated by cutting the sciatic nerve. Starting either 7 days or 1 month after the surgery, the animals were trained on a treadmill. Two months after denervation, the soleus muscle was investigated using light and electron microscopy and biochemical methods. Control soleus muscles were obtained from non-trained animals: the untreated and the 2-month denervated. RESULTS: Locomotor training caused slight increase in denervated rat soleus muscle weight and significant increase in its fiber diameter. The training positively affected some of the factors that were believed to be the reasons of atrophy irreversibility, because of significant increase in the number of capillary blood vessels and muscle fiber nuclei with the concomitant decrease in the number of severely damaged muscle fibers and amount of collagen. Morphology of the contractile apparatus was also improved as more regular organization of sarcomeres and the hexagonal arrangement of myosin filaments was evident. Moreover, the amount of myosin heavy chains (MHC) significantly increased after training. The effects were more evident in the animals with longer training. CONCLUSION: Passive movement seems to attenuate some of the pathologic processes within the denervated muscle.     

Tenascin-C in the cochlea of the developing mouse

Tenascin-C is a glycoprotein of the extracellular matrix that acts in vitro as both a permissive and a nonpermissive substrate for neurite growth. We analyzed, by immunocytochemistry, the distribution of tenascin-C along neural growth pathways in the developing mouse cochlea. In the spiral lamina, tenascin-C coexists in a region where nerve bundles arborize. In the organ of Corti, tenascin-C lines the neural pathways along pillar and Deiters' cells before and during the time of nerve fiber ingrowth. By embryonic day 16, tenascin-C is abundant on the pillar side of the inner hair cell but does not accumulate on the modiolar side until about birth, a time after the arrival of afferent fibers. The synaptic zones beneath outer hair cells are strongly labeled during the time when early events in afferent synaptogenesis are progressing but not during the time of efferent synaptogenesis. At the age when most neural growth ceases, tenascin-C immunoreactivity disappears. Faint tenascin-C immunolabeling of normal hair cells, strong tenascin immunolabeling in pathological hair cells of Bronx waltzer (bv/bv) mice, and staining for beta-galactosidase, whose gene replaces tenascin in a "knockout" mouse, indicate that hair cells supply at least part of the tenascin-C. The changing composition of the extracellular matrix in the synaptic region during afferent and efferent synaptogenesis is consistent with a role for tenascin in synaptogenesis. The presence of tenascin-C along the growth routes of nerve fibers, particularly toward the outer hair cells, raises the possibility that growth cone interactions with tenascin-C helps to guide nerve fibers in the cochlea.     

Prevention of denervation atrophy in muscle: Mammalian neurotrophic factor is not transferrin

Atrophy in a denervated muscle results from the disuse caused by paralysis of the muscle, and from the loss of special neurotrophic substances. Daily injections of proteins extracted from rats' sciatic nerves have been shown to prevent the non-disuse atrophy of rats' muscles denervated for 7 days. The trophic factor from chicken sciatic nerve which stimulates differentiation in aneural chick muscle in vitro has been purified and found to be closely similar to transferrin. We undertook to determine whether the trophic properties of mammalian nerve extract on denervated rats' muscles in vivo were due to the presence of serum transferrin in the extract. Atrophy was measured as the reduction in cross-sectional areas of type IIB fibers in the extensor digitorum longus muscle. Muscles denervated for 7 days and injected daily with 1 of several doses of iron-conjugated rat transferrin exhibited a rate of atrophy equivalent to that in denervated muscles that either were not treated or were injected with saline. Denervated muscles injected with crude extract of rats' sciatic nerves had significantly less atrophy than their controls. Removal of transferrin from the crude extract by immunoaffinity chromatography did not diminish its ameliorative effects on denervated muscle. Therefore, the trophic action of mammalian nerve extract on denervated rats' muscles in vivo is not due to the presence of serum transferrin in the extract.     

Treatment of denervated muscle by gangliosides

Short-duration cooling of the nerve to the extensor digitorum longus muscle of the rat in vivo induced partially reversible denervation of the muscle and atrophy in the type 2 muscle fibers. Increases in cyclic adenosine monophosphate, cyclic guanosine monophosphate phosphodiesterase, adenylate cyclase, and guanlate cyclase were observed in the denervated muscle. Treatment with gangliosides of the bovine brain cortex seemed to improve the excitability of the surviving motor units and to encourage recovery of neuromuscular trophic control, but it did not affect the nerve conduction velocity or the contractile properties of the denervated muscle.     

Effects of tenascin-C in cultured hippocampal astrocytes: NCAM and fibronectin immunoreactivity changes

Tenascin-C is an extracellular matrix glycoprotein with trophic and repulsive properties on neuronal cells, involved in migratory processes of immature neurons. Previous reports demonstrated that this molecule is produced and secreted by astrocytes, in vitro after activation by bFGF or in vivo after CNS lesion. In injured brain the expression of tenascin-C has been correlated with the glial reaction since it was observed in regions suffering a dramatic glial proliferation and hypertrophy. In this report we show that the treatment of cultured hippocampal astrocytes with tenascin-C results in an increased fibronectin and NCAM immunoreactivities. In addition, treated astrocytes form longer extensions than control ones. The number of cells as well as the levels of GFAP mRNA and protein immunoreactivity are not modified after tenascin-C treatment. The present changes may, therefore, be related to the modification of the adhesive properties of astrocytes to the substrate. These observations are compatible with the hypothesis that tenascin-C may contribute to the glial scarring process. GLIA 20:231–242, 1997. © 1997 Wiley-Liss, Inc.     

Properties of medial gastrocnemius motor units and muscle fibers reinnervated by embryonic ventral spinal cord cells

Severe muscle atrophy occurs after complete denervation. Here, Embryonic Day 14-15 ventral spinal cord cells were transplanted into the distal tibial nerve stump of adult female Fischer rats to provide a source of neurons for muscle reinnervation. Our aim was to characterize the properties of the reinnervated motor units and muscle fibers. Some reinnervated motor units contracted spontaneously. Electrical stimulation of the transplants at increasing intensity produced an average (+/- SE) of 7 +/- 1 electromyographic and force steps. Each signal increment represented the excitation of another motor unit. These reinnervated units exerted an average force of 12.0 +/- 1.5 mN, strength similar to that of control fatigue-resistant units. Repeated transplant stimulation depleted 17% of the muscle fibers of glycogen, an indication of some functional reinnervation. Reinnervated (glycogen-depleted), denervated (no cells transplanted), and control fibers were of histochemical type I, IIA, or IIB. Fibers of the same type were grouped after reinnervation. The proportion of fiber types also changed. Reinnervated fibers were primarily type IIA, whereas most fibers in denervated and control muscles were type IIB. Reinnervated fibers of each type had significantly larger cross-sectional areas than the corresponding fiber types in denervated muscles. These data suggest that neurons with different properties can reside in the unusual environment of the adult rat peripheral nerve, make functional connections with muscle, specify muscle fiber type, and reduce the amount that each type atrophies.     

Selective type II muscle fiber hypertrophy in severe infantile spinal muscular atrophy

The diagnostic muscle biopsy finding in severe infantile spinal muscular atrophy (Werdnig-Hoffmann disease, SMA type 1) is considered to be large-group atrophy with isolated clusters of hypertrophic type I myofibers. We present a unique case of severe infantile spinal muscular atrophy with selective hypertrophy of type II myofibers. A male infant presented at age 2 months with breathing difficulties and by age 4 months was hypotonic and weak. Electromyography revealed denervation in all extremity muscles, and nerve conduction velocities were normal but with small compound muscle action potentials. Quadriceps muscle biopsy revealed many hypertrophied type II myofibers (myofibers with a mean least diameter of 25.4 microns). In contrast, the largest type I myofibers were 20 microns in least diameter (mean diameter, 14.9 microns), and there was a normal-size population of type II fibers (mean diameter, 15.7 microns). In addition, sheets of atrophic type I and type II fibers averaged 2.0 microns in least diameter. Sural nerve biopsy was normal. Breathing difficulties progressed, with death ensuing at age 5 1/2 months. Autopsy revealed atrophy of ventral spinal roots with normal dorsal roots. There was loss of anterior horn cells, while remnant neurons were reduced in size. No other pathologic changes were identified. This case indicates that in severe infantile spinal muscular atrophy, relative sparing of the motor units with type II myofibers may occur.