Denervation of rat soleus muscle and simultaneous administration of high doses of corticosteroids for 7 days caused marked muscle fiber atrophy and selective loss of thick myofilaments from many muscle fibers by light and electron microscopy. Myosin heavy chain/ actin ratios were greatly reduced on polyacrylamide gel electrophoresis. Nerve crush instead of cut permitted reinnervation after 2 weeks and demonstrated the reversibility of the muscle changes within a week after reinnervation. There was formation of new thick filaments and their reintegration into myofibrils without further breakdown, although large areas of Z-disc streaming appeared. The mechanism of A-band breakdown remains obscure, but it presumably starts with limited proteolysis and continues with disaggregation of myosin molecules. This is consistent with our observation that the muscle fibers retain a relatively good reactivity to antibodies against myosin heavy chain 1 week after denervation and corticosteroid administration. A syndrome recalling these experiments is seen in severely asthmatic patients receiving corticosteroids and pharmacologically paralyzed for mechanical respiration. 0 1992 John Wiley & Sons, Inc.
Key words: corticosteroids denervation reinnervation myosin thick myofilaments rat soleus muscle proteolysis MUSCLE & NERVE 15:1290-1298 1992
LOSS AND RENEWAL OF THICK MYOFILAMENTS IN GLUCOCORTICOI DTREATED RAT SOLEUS AFTER DENERVATION AND REINNERVATION ROBERTO MASSA, MD, STIRLING CARPENTER, MD, PAUL HOLLAND, PhD, and GEORGE KARPATI. MD
Clinical or experimental administration of large amounts of glucocorticoids causes significant atrophy of skeletal muscles.”2”0”“””2*Synthesis of all major myofibrillar proteins is impaired and their catabolism is increased. 13,14,20 This also happens in the more marked atrophy that follows denervation. l 7 The combination of high-dose dexamethasone and denervation (DEX-DEN) has been shown to cause greater atrophy in rat skeletal muscles than either intervention alone.’“ But, in addition to its effects on the general turnover of
From the Department of Neurology-Neurosurgery, McGill University, and The Montreal Neurological Institute, Montreal, Quebec, Canada Acknowledgments Supported by the Medical Research Council of Canada and the Muscular Dystrophy Association of Canada. Dr. Massa was a postdoctoral research Fellow of Unione ltaliana Lotta alla Distrofia Musculare. Rome, Italy. We thank Claude Guerin for his skillful preparation of the SDS gels and immunoblots and Dr. Donald Fischman, MD. for his gift of the monoclonal antibodies (MF18 and MF20) to myosin heavy chains Address reprint requests to Dr Stirling Carpenter, Montreal Neurological Institute, 3801 University Street, Montreal, Quebec, Canada H3A 284. Accepted for publication April 28, 1992 CCC 0148-639X/92/111290-09 0 1992 John Wiley & Sons, Inc.
1290
Renewal of Thick Myofilaments
myofibrillar proteins, DEX-DEN causes an unusual reaction: preferential depletion of thick myofilaments as shown by electron microscopy, and of myosin as shown by electrophoretic gels.” ’This has never been seen after dexamethasone alone. We have not seen it after denervation alone, nor does it appear in most accounts of the atrophy of denervation. The cellular mechanisms of this change are not clear. It has not been shown whether the lesions are reversible and, if so, how. The present series of experiments on the soleus muscles of DEX-DEN rats was designed to clarify these points. T o evaluate the reversibility of the lesions, we studied temporarily denervated DEXDEN soleus muscles at various intervals before and after denervation and reinnervation.
MATERIALS AND METHODS Experimental Design. Guidelines of The Canadian Council on Animal Care were followed. Adult (300 g) male Sprague-Dawley rats were anaesthetized with pentobarbital (50 mg/kg, IF). In the left hind limb, irreversible denervation
MUSCLE & NERVE
November 1992
(DEX-DEN-IRR) of soleus muscle was effected by removing 10 mm of the sciatic nerve at the sciatic notch. On the right, reversible denervation (DEXDEN-R) of the soleus was effected by crushing the posterior tibial nerve at the popliteal fossa for 20 seconds with a jeweller's forceps. T h e effectiveness of the crush in producing denervation was assessed in each animal by the absence of mechanical response of the leg muscles after supramaximally stimulating the posterior tibial nerve proximal to the crush level through a unipolar platinum ball electrode. T h e stimuli were single, supramaximal square waves of 0.7 ms duration, 100 Hz frequency, and 0.3- 1.0 V amplitude (Grass S4 CR stimulator). T h e same procedure was employed just before killing the animals for tissue removal, to assess whether reinnervation of the soleus had occurred in the right hind limb (the DEX-DEN-R model). Dexamethasone (1 mg/mL) was injected daily subcutaneously in the paraspinal region at a dose of 5 mg/kg of body weight for 7 consecutive days starting on the day of denervation. Intact control animals were injected with saline solution. T h e animals were killed by pentobarbital overdose (135 mg/kg, IP) at different time intervals after denervation, and the solei from both hind limbs totally removed. Preliminary experiments established that a l-week interval between denervation and sacrifice was optimal for observing severe depletion of thick myofilaments. Pilot studies also established that reinnervation in the DEX-DEN-R muscles occurred approximately 10 days after denervation by tibial nerve crush. DEX-DEN rats were killed 1 week ( n = 4), 2 weeks (n = 7), 3 weeks (n = 6 ) , 4 weeks ( N = 6 ) , 5 weeks (n = 6 ) and 7 weeks ( n = 3) after denervation. Two intact control rats were killed at each time interval. When stimulated before sacrifice, none of the DEX-DEN muscles showed signs of reinnervation on either size at 1 week. By 2 weeks after tibial nerve crush (DEX-DEN-R), all animals showed good plantar flexion of the right foot to stimulation, but no evidence of reinnervation on the DEX-DEN-IRR side. Statistical analysis of muscle wet weight was performed by the ANOVA test. P < 0.05 was considered significant. Half of each soleus muscle split in the middle along the longitudinal axis was removed in an isometric clamp, fixed in 2% glutaraldehyde in cacodylate buffer, weighed, and processed for resin embedding. T h e other half was weighed, then frozen in isopentane chilled with liquid nitrogen for light microscopy
and biochemistry. From the midpoint of the muscles, 8 km thick, transverse cryostat sections were reacted for myosin ATPase activity after preincubation at pH 9.4, 4.6, and 4.2. For immunocytochemistry, two monoclonal antibodies (Ab) to epitopes in the S1 (MF18) and LMM (MF20) regions of MHC were used4 (generous gift from D. Fischman, MD, Cornell University). Four-micrometer thick, transverse cryostat sections were fixed in acetone (1 minute), incubated in the undiluted primary Ab (30 minutes) followed by a biotinylated goat antimouse IgG Ab (30 minutes), which was then visualized by the streptavidin- peroxidase technique. On control sections, incubation of the primary Ab was substituted by incubation in 2% normal goat serum. Transverse and longitudinal semithin resin sections, stained with paraphenylene diamine, were examined by phase microscopy. Ultrathin sections, stained with uranyl acetate and lead citrate, were examined with an electron microscope.
''
From the animals killed 1, 2, 3, and 6 weeks after crush denervation, and from the control animals, an aliquot of frozen muscle from the right soleus was used for biochemistry. For one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) and Western blot analysis, 10 mg of frozen muscle from each of 3 or more animals of the same group were taken and pooled. Samples were homogenized and rotein was determined as previously described.' Samples were diluted with SDS sample buffer to obtain a final protein concentration of 0.5 to 1.5 mg/L. SDS-PAGE was performed according to the method of Laemmli,15 using a 12% acrylamide separating gel, loading
Biochemical Techniques.
Microscopic Techniques.
Renewal of Thick Myofilaments
1
2
3
5
4
7
Weeks After Denervation
FIGURE 1. Mean wet weight with standard deviations of soleus muscles at weekly intervals after denervation. Dexamethasone was given for the first 7 days to all animals except controls.
MUSCLE & NERVE
November 1992
1291
1.5 pg of protein per sample on a miniprotein cell (Biorad Lab, Mississauga, Ontario, Canada). Molecular weight markers were run on the same gel. The gel was stained with Coomassie blue and scanned at 550 nm using a Zeiss spectrophotometer (Model PMQII, Carl Zeiss Ltd., Oberkochen, Germany). The peak areas of actin and myosin heavy chain (MHC) bands were measured using a MOP3 image analyzer (Carl Zeiss Ltd.) and the MHC/actin ratio was calculated by taking the sum of the number of pixels times their intensity from 1 to 256. For Western blot analysis of MHC immunoreactivity, a replicate gel was electroblotted for 16 hours at 25 V in a minitransblot apparatus (Biorad) in transfer buffer (tris/glycine/20% MeOH). The blot was blocked for 45 minutes at 37°C with 10% skim milk in tris-buffered saline with Tween." The MF20 monoclonal Ab, which reacts with all MHC isoforms, was then applied at a dilution of 1:7 for 1 hour at room temperature. An antimouse IgG, alkaline-phosphatase-conjugated secondary Ab was used for visualization. RESULTS
The changes in wet weight of soleus muscles in the different experimental conditions and at different time intervals are shown in Figure 1. On the side of nerve excision (DEXDEN-IRR), the average wet weight was significantly and progressively reduced, compared with Muscle Weight.
intact controls, throughout the experiment, reaching a minimum plateau by 5 weeks. On the side of nerve crush (DEX-DEN-R), the average wet weight was reduced to approximately the same extent, at both 1 and 2 weeks. At 3, 4, and 5 weeks, the average weight, though progressively increasing, was still significantly lower than that of matched intact controls. By 7 weeks, this difference was no longer significant. The solei studied 1 week after denervation showed variability of staining for myosin ATPase at all pHs. Many fibers had surprisingly well-preserved activity, while others, regardless of fiber type, had a decreased level (Fig. 2A and B). There was conspicuous variability of staining on transverse view within single fibers. N o difference was noted between the two modalities of denervation. In DEX-DEN-R muscles studied after 3 or more weeks, the ATPase activity was normal. The MF18 Ab gave preferential staining of type I1 extra- and intrafusal fibers in control muscles, whereas MF20 gave an equal staining of all fiber types (Figs. 3A and 4A). With both Abs, the staining had a clear myofibrillar pattern. There was irregular reduction in the intensity of staining at 1 week after denervation in many muscle fibers compared with controls (Figs. 3 R and 4B). Semithin and ultrathin sections at 7 and 14 days were similar in the muscles with either Microscopic Studies.
B
A
FIGURE 2. Myosin ATPase reactions, preincubation pH 9.4. Compared with control (A), l-week DEX-DEN muscle ( 8 )shows a moderate general reduction in activity with marked focal or diffuse loss in some fibers (examples marked by arrows) (x350).
1292
Renewal of Thick Myofilaments
MUSCLE & NERVE
November 1992
A
6
FIGURE 3. Monoclonal antibody to MHC (MF18) which preferentially stains type II fibers. Compared with control (A), 1-week DEX-DEN muscle (B),shows marked focal reduction in the staining of many type II fibers (x300).
crushed or cut nerves. Varying degrees of loss of A-band density affected 40%-100ojO of fibers on individual semithin sections. The degree of loss at 14 days did not exceed that seen at 7 days, although the size of muscle fibers was smaller. The mitochondrial pattern became profoundly disturbed. On transverse sections, the normal gridlike mitochondrial densities localized to the I band
were replaced by diffusely scattered punctate densities. The rounded mitochondria normally seen aggregated beneath the sarcolemma of type I fibers were largely absent at 7 days and almost totally absent at 14 days. On longitudinal sections, mitochondria appeared elongated in the long axis of the fibers and frequently had migrated into the A band.
B
A
FIGURE 4. Antibody to MHC (MF20) which stains myofibrils of all fibers uniformly in control (A). One-week DEX-DEN muscle (6) shows less intense stain in several fibers (~300).
Renewal of Thick Myofilaments
MUSCLE & NERVE
November 1992
1293
B
A
FIGURE 5. Three weeks postdenervation. In muscle with nerve cut (A), a central fiber shows absence of A-band density. Z discs are out of register. The other fibers show mild A-band pallor and, in the fiber on the bottom, mitochondria have migrated into the A band from their usual para-Z-disc position. (B) In the reinnervated muscle, the density of A bands is normal except where sarcomeres are obliterated by Z-disc streaming, which covers a large zone, 2,in the fiber below, and is multifocal in the fiber above. Subsarcolemmal myofibril-freezone contains large dark mitochondria (arrow). Resin sections, paraphenylene diamine, phase optics ( ~ 7 5 0 ) .
After 14 days in DEX-DEN-IRR muscles, the packing density of mitochondria, as seen on transverse sections, lessened despite further shrinkage of cell girth. A band loss persisted (Fig. 5A). Marked loss of thick myofilaments was found by electron microscopy in many muscle fibers (Fig. 6A). Loss of myofilaments did not appear to exceed loss of cell girth. In fibers with extensive A-band loss, the alignment of Z discs became disturbed, but even at 7 weeks there did not appear to be any foci of total myofibrillar breakdown. Z-disc streaming was almost totally absent from DEX-DEN-IRR muscles. In DEX-DEN-R solei at 21 days, there was persistent A-band loss only in a few fibers of small size, but the majority of fibers showed either normal A-band density or extensive areas of Z-disc streaming (Figs. 5B, 6B, and 7). Many of these fibers had prominent subsarcolemmal areas that lacked myofibrils but contained large mitochondria. Such areas were never seen in fibers with A-band loss, whether focal or diffuse. At further time intervals, as the fibers regained normal size, these areas tended to be less and less prominent. Large areas of Z-disc streaming persisted; by 7 weeks they were almost the only abnormality. Foci of complete myofibrillar breakdown, signaled by lack of Z-disc density were rare and seen in only 2 specimens at 4 weeks.
1294
Renewal of Thick Myofilarnents
Biochemical Studies. One-dimensional SDSPAGE showed substantial preferential depletion of MHC relative to total protein in DEX-DEN-R muscles studied 1 week after crush denervation (Fig. 8). Because no consistent depletion of actin was evident in the electrophoretic gels, we chose MHC/actin ratio as a relative measure for MHC depletion. MHC/actin ratio was 1.5 to 1 in control solei, and it dropped to 0.78 to 1 at 1 week postdenervation (Table 1). In reinnervated muscles, by 3 weeks postdenervation, it regained a normal value (1.6 to l), which persisted at 6 weeks (1.62 to 1). No bands of abnormal molecular weight were observed in the denervated or reinnervated muscles. Three bands between MHC and glycogen phosphorylase appeared significantly paler in the 2-week DEX-DEN muscle than in the control. These were in the approximate expected location of myomesin (185 kd),6,25 M protein (165 kd), and C protein (145 kd),'8,25 but exact identification of these bands is not possible without appropriate antibodies. Western blot analysis showed that MF20 Ab immunoreactivity was localized mainly to the band corresponding to MHC in all the experimental conditions tested (not illustrated). A titer for the Ah (1:7) was chosen to allow detection of small amounts of MHC proteolytic fragments in the DEX-DEN muscles. However, such subfragments
MUSCLE & NERVE
November 1992
A
B
FIGURE 6. Three weeks postdenervation. (A) Portions of three fibers whose relative pallor indicates thick filament loss in DEX-DENIRR. Mitochondria form small random clusters. (B)Reinnervated fibers have normal myofibrils and subsarcolemmal areas with very large mitochondria, ribosomes, and glycogen. DEX-DEN-R. [Both (A) and (6)~6750.1
were not found, though overloading of the blot resulted in some diffusion of the staining to adjacent lanes and reactivity of several small bands between 80 and 100 kd in all the lanes.
DISCUSSION
In the DEX-DEN model, the combination of denervation and high dose dexamethasone for 7 days induced the following effects: (1) a reduction of the muscle wet weight by more than one third; (2) a marked reduction (-44%) of the MHC/actin ratio, revealed by SDS-PAGE; and (3) a severe preferential depletion of thick myofilaments revealed by EM. These findings are similar to those reported by Rouleau et al.,lg except that severe depletion of thick myofilaments was consistently achieved within 7 days by using higher doses of dexamethasone. The temporary denervation produced by nerve crush proved as effective in producing these changes as nerve excision. Nerve crush was always
Renewal of Thick Myofilaments
followed by reinnervation within 2 weeks. Muscle reinnervation and the withdrawal of dexamethasone resulted in rapid reconstitution of normal sarcomere pattern in all animals. By 3 weeks postdenervation, most myofibers contained normally dense A bands. Evidence that these fibers had previously been denervated was the presence of elongated mitochondria protruding into the A bands. The myosin-actin ratio returned to normal after 3 weeks, although the wet weight of the muscle was not normal until 7 weeks. Withdrawal of dexamethasone was not sufficient, by itself, for the reconstitution of thick myofilaments in permanently denervated solei. Reinnervation was an essential condition for reversibility. In most accounts of denervation, thick and thin filaments are lost together. 1922 The exception to this is from reports by Jakubiec-Puka et a1.8*Y The actin-myosin ratio dropped from 1.78 to 0.80 after 27 days. Reinnervation by 20 days after sciatic nerve crush in their experiments resulted in a relative increase in myosin filaments in all fibers,
MUSCLE 8, NERVE
November 1992
1295
FIGURE 7 . Higher magnification of Figure 6B. Note the normal myofibrils and the large myofibril-free space which contains large mitochondria, glycogen, and ribosomes ( ~ 5 0 , 0 0 0 ) .
and a return toward the regular hexagonal arrangement of thin filaments around each thick one. Myofibrils became normal in appearance in the center of fibers, while around the periphery they were small and irregular. T h e actin-myosin ratio returned to normal by 45 days. These findings after simple denervation were not confirmed by Rouleau et al.") Breakdown of actin in dener-
Table 1. Computer image analysis of gels of control and DEX-DEN-R muscles.
Control 1 week 2 weeks 3 weeks 6 weeks
Myosin
Actin
MIA
21 1494 141a25 236325 21 8757 240293
140887 180831 163001 13691 1 146583
1.50 0.78 1.45 1.60 1.64
Myosin and actin bands were circumscribed and a histogram taken of each band Data are composed of the sum of the number of pixels times their intensity on a scale of 1 to 256 in columns
1296
Renewal of Thick Myofilaments
vated muscle appears to occur by a nonlysosomal, non- Ca++-dependent,ATP-requiring system.5 Our histochemical and microscopic immunocytochemical findings were somewhat unexpected. On the basis of the EM and biochemical results, a major loss of microscopic myosin ATPase reactivity and of MHC immunoreactivity would be expected by 7 days in the majority of muscle fibers. The retention to a considerable degree of these activities could be explained if the disappearance of thick myofilaments were due to their simple disassembly into myosin monomers, which remained as a soluble cytoplasmic pool retaining ATPase activity. Myosin molecules can be disassembled in vitro from isolated thick myofilaments or from intact cells by exposure to high ionic strength solutions.' Maw and Rowe have succeeded in first solubilizing myosin and subsequently reconstructing thick myofilaments in intact cells by returning the bathing salt concentration to normal while preventing egress of the solubilized myosin molecules.16 In our model, it is
MUSCLE & NERVE
November 1992
- 92.5 - 69
- 46 1
2
3
4
5
FIGURE 8. One-dimensional SDS-PAGE stained with Coomassie blue. Lane 1: Control. Lane 2: DEX-DEN-R at 1 week with no reinnervation as yet. Lane 3: DEX-DEN-R at 2 weeks after crush with reinnervation. Lane 4: DEX-DEN-R at 3 weeks after crush with reinnewation. Lane 5: DEX-DEN-R at 6 weeks after crush with reinnervation. Molecular weights are indicated on right. Arrow indicates MHC which shows marked diminution in lane 2 and gradual recovery in lanes 3, 4, and 5. Arrowheads indicate other bands which become less dense after DEXDEN and may represent myomesin, M protein, and C protein.
noreactivity. We did not, however, detect partial proteolytic fragments by Western blot analysis. On the basis of our results, we suggest that, in the DEX-DEN model, limited proteolysis of MHC (possibly at the hinge region of the molecule) occurs first. The truncated myosin monomers then disaggregate and subsequently undergo total catabolism. The amount of remaining disaggregated, but not yet fully catabolized, MHC residues is probably sufficient to account for the microscopic partial ATPase activity and MHC immunoreactivity, in view of the degree of shrinkage occurring in the muscle fibers. Proteolysis is known to be increased in denervated m ~ s c l e ~ ~ and ” ~in* ~ steroid-treated The preferential susceptibility of MHC in the DEX-DEN model remains unexplained. The thick myofilament depletion is rapidly and fully reversible upon reinnervation of the muscle and discontinuation of the drug. For this reconstitution, de novo synthesis of MHC is probably necessary. Human disease analogous to our experiments has been reported in patients given high-dose corticosteroids while receiving pharmacologic neuromuscular blockade to facilitate artificial respiration, usually for status a s t h m a t i c u ~ .T~h*e~duration ~ of‘ neuromuscular blockade in 2 1 patients varied from 6 to 40 days. After cessation they remained quadriparetic. Recovery was generally complete over weeks to months. Biopsies in at least 3 patients showed extensive selective loss of thick myofilaments. 3 2 3
REFERENCES
possible that corticosteroids would raise the intracellular ionic concentration and also possibly that this effect would be magnified by the cell shrinkage that accompanies denervation,22 but we have no direct evidence for this. It seems unlikely that the ionic concentration needed to disassemble thick filaments into myosin monomers could be reached in vivo. If thick myofilaments had only been disassembled into individual myosin monomers, which were still present, this would account for the microscopic myosin ATPase activity and MHC immunoreactivity persisting at 1 week, but it would not be consistent with the degree of loss of MHC seen on SDS-PAGE. On the other hand, limited proteolysis of MHC would account for the SDSPAGE findings and be consistent with the persistence of myosin ATPase activity and MHC immu-
Renewal of Thick Myofilaments
1. Braunstein PW Jr. Girolami V: Experimental corticosteroid myopathy. Acta Neuropathol (Berl) 1981;55:167- 172. 2. Clarke AF, Vignos PJ Jr: Experimental corticosteroid myopathy: effect on myofibrillar ATPase activity and protein degradation. Mwcle Nerve 1979;2:205-273. 3. Danon MJ, Carpenter S: Myopathy with thick filament (myosin) loss following prolonged paralysis with vecuronium during steroid treatment. Muscle Nerve 1991; 14: 1131-1139. 4. Fischman DA, Masaki T: Immunochemical analysis of myosin with monoclonal antibodies, in Anderson W, Sadler W (eds): Perspectives in Dfferentiation and Hypertrophy. Amsterdam, Elsevier, 1980, pp 279-291. 5. Furono A, Goodman MN, Goldbert HL. Role of different systems in the degradation of muscle proteins during denervation atrophy. J Biol Chem 1990;265:8550-8557. 6. Grove BK, Kurer V, Lehner C, Doetschman TC, Perriard J-C, Eppenberger HM: A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies. J Cell Biol 1984;98:518-524. 7. Harrington WF, Rodgers ME: Myosin. Ann Rev Biochrm 1984;53:35-73. 8 . Jakubiec-Puka A, Kulesza-Lipka D, Krajewski K: The con-
MUSCLE & NERVE
November 1992
1297
tractile apparatus of striated muscle in the course of atrophy and regeneration: I. Myosin and actin filaments in the denervated rat soleus. Cell ?'issue Res 1981;220:65 1-663. 9. Jakubiec-Puka A, Kulesza-Lipka D, Kordowska J : The contractile apparatus of striated niusclc in the course of atrophy and regeneration: 11. Myosin and actin lilaments in mature rat .solcus muscle regenerating after reinnervation. Cell Ti.rsue Res 1982;227:641-650. 10. Jirmanova I, Soukup T , Zelena J : T h e pathomorphology of developing skeletal muscles of rabbits treated with glucocorticoids. Virchow's Arch (Cell Pathol) 1982;38: 323-335. 11. Johnson DA, Gautsch JW, Sportsman JR, Elder JH: Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose. Gene Anal Techniq 1984;1:3-8. 12. Karpati G, Pouliot Y, Carpenter S: Expression of immunoreactive major histocompatibility complex products in human skeletal muscles. A n n Neurol 1988;23:64-72. 13. Kelly FJ, Goldspink DF: T h e differing responses of four muscle types to dexamethasone treatment in the rat. Bzochen J 1982;208:147-151. 14. Kelly FJ, McGrath JA, Goldspink DF, Cullen MJ: A morphologicalbiochemical study on the actions of corticosteroids on rat skeletal muscle. Muscle Nerve 1986;9:1- 10. 15. Laemmli VK: Cleavage and structural proteins during the assembly of the head of bacterophage T4. Nature 1970;227:680-685. 16. Maw MC, Rowe HJ: T h e reconstitution of myosin filaments in rabbit psoas muscle from solubilized myosin. J Muscle Re.( Cell Motil 1986;7:97- 109. 17. Metafora S, Felsani A, Cotrufo K, Tajana GF, Dilorio G , Del Rio A, DePrisco PP: Neural control of gene expression in skeletal muscle fibers: the nature of the lesion in the
1298
Renewal of Thick Myofilarnents
muscular protein-syrithetizing machinery following denervation. Proc R SOCLond ( B ) 1980;209:239-273. 18. Obinata T, Maruyama K, Sugita H, Kohoma K, Ebashi S: Dynamic aspects of structural proteins in vertebrate skeletal muscle. M U X ~Nerve P 1%1;4:456-488. 19. Rouleau G , Karpati G, Carpenter S, Soza M, Prescott S, Holland P: Glucocorticoid excess induces preferential depletion of myosin in denervated skeletal muscle fibers. Muscle Nerve 1987;10:428-438. 20. Shoji S, Pennington RJT: T h e effect of cortisone on protein breakdown and synthesis in rat skeletal muscle. Mol Cell Enducrinol 1977;6: 159- 169. 21. Stern LZ, Fagan JM: 'I'he endocrine myopathies, in Vinken DJ, Bruyn GW (eds): Diseases of Muscles, Handbook of Clinical Neurology. Amsterdam, North Holland, 1979, vol 4, pp 235-258. 22. Stonnington, M M , Engel AG: Normal and denervated muscle: a morphometric study of fine structure. Neurology (Minneap) 197 3 ;23 :7 14- 724. 23. Waclawik AJ, Sufit RL, Beinlich BR, Schutta HS: Acute myopathy with selective degradation of myosin filaments following status asthmaticus. J Neurol Scz 1990;98 (suppl):470. 24. Walsh G, DeVivo D, Olson W: Histochemical and ultrastructural changes in rat muscle. Occurrence following adrenal corticotropic hormone, glucocorticoids and starvation. Arch Neurol 1972;24:83-93. 25. Wang K: Sarcomere-associated cytoskeletal lattices in striated muscle, in Shag JW (ed): Cell and Musrle Motility. New York, Plenum, 1985, vol 6, p p 315-369. 26. Weinstock Im, Iodice AA: Acid hydrolase activity in muscular dystrophy and denervation atrophy, in Dingle JT, Fell FB (eds): Lysosomes in Biology and Pathology. New York, Elsevier, 1969, p p 450-468.
MUSCLE & NERVE
November 1992
Key words: corticosteroids denervation reinnervation myosin thick myofilaments rat soleus muscle proteolysis MUSCLE & NERVE 15:1290-1298 1992
LOSS AND RENEWAL OF THICK MYOFILAMENTS IN GLUCOCORTICOI DTREATED RAT SOLEUS AFTER DENERVATION AND REINNERVATION ROBERTO MASSA, MD, STIRLING CARPENTER, MD, PAUL HOLLAND, PhD, and GEORGE KARPATI. MD
Clinical or experimental administration of large amounts of glucocorticoids causes significant atrophy of skeletal muscles.”2”0”“””2*Synthesis of all major myofibrillar proteins is impaired and their catabolism is increased. 13,14,20 This also happens in the more marked atrophy that follows denervation. l 7 The combination of high-dose dexamethasone and denervation (DEX-DEN) has been shown to cause greater atrophy in rat skeletal muscles than either intervention alone.’“ But, in addition to its effects on the general turnover of
From the Department of Neurology-Neurosurgery, McGill University, and The Montreal Neurological Institute, Montreal, Quebec, Canada Acknowledgments Supported by the Medical Research Council of Canada and the Muscular Dystrophy Association of Canada. Dr. Massa was a postdoctoral research Fellow of Unione ltaliana Lotta alla Distrofia Musculare. Rome, Italy. We thank Claude Guerin for his skillful preparation of the SDS gels and immunoblots and Dr. Donald Fischman, MD. for his gift of the monoclonal antibodies (MF18 and MF20) to myosin heavy chains Address reprint requests to Dr Stirling Carpenter, Montreal Neurological Institute, 3801 University Street, Montreal, Quebec, Canada H3A 284. Accepted for publication April 28, 1992 CCC 0148-639X/92/111290-09 0 1992 John Wiley & Sons, Inc.
1290
Renewal of Thick Myofilaments
myofibrillar proteins, DEX-DEN causes an unusual reaction: preferential depletion of thick myofilaments as shown by electron microscopy, and of myosin as shown by electrophoretic gels.” ’This has never been seen after dexamethasone alone. We have not seen it after denervation alone, nor does it appear in most accounts of the atrophy of denervation. The cellular mechanisms of this change are not clear. It has not been shown whether the lesions are reversible and, if so, how. The present series of experiments on the soleus muscles of DEX-DEN rats was designed to clarify these points. T o evaluate the reversibility of the lesions, we studied temporarily denervated DEXDEN soleus muscles at various intervals before and after denervation and reinnervation.
MATERIALS AND METHODS Experimental Design. Guidelines of The Canadian Council on Animal Care were followed. Adult (300 g) male Sprague-Dawley rats were anaesthetized with pentobarbital (50 mg/kg, IF). In the left hind limb, irreversible denervation
MUSCLE & NERVE
November 1992
(DEX-DEN-IRR) of soleus muscle was effected by removing 10 mm of the sciatic nerve at the sciatic notch. On the right, reversible denervation (DEXDEN-R) of the soleus was effected by crushing the posterior tibial nerve at the popliteal fossa for 20 seconds with a jeweller's forceps. T h e effectiveness of the crush in producing denervation was assessed in each animal by the absence of mechanical response of the leg muscles after supramaximally stimulating the posterior tibial nerve proximal to the crush level through a unipolar platinum ball electrode. T h e stimuli were single, supramaximal square waves of 0.7 ms duration, 100 Hz frequency, and 0.3- 1.0 V amplitude (Grass S4 CR stimulator). T h e same procedure was employed just before killing the animals for tissue removal, to assess whether reinnervation of the soleus had occurred in the right hind limb (the DEX-DEN-R model). Dexamethasone (1 mg/mL) was injected daily subcutaneously in the paraspinal region at a dose of 5 mg/kg of body weight for 7 consecutive days starting on the day of denervation. Intact control animals were injected with saline solution. T h e animals were killed by pentobarbital overdose (135 mg/kg, IP) at different time intervals after denervation, and the solei from both hind limbs totally removed. Preliminary experiments established that a l-week interval between denervation and sacrifice was optimal for observing severe depletion of thick myofilaments. Pilot studies also established that reinnervation in the DEX-DEN-R muscles occurred approximately 10 days after denervation by tibial nerve crush. DEX-DEN rats were killed 1 week ( n = 4), 2 weeks (n = 7), 3 weeks (n = 6 ) , 4 weeks ( N = 6 ) , 5 weeks (n = 6 ) and 7 weeks ( n = 3) after denervation. Two intact control rats were killed at each time interval. When stimulated before sacrifice, none of the DEX-DEN muscles showed signs of reinnervation on either size at 1 week. By 2 weeks after tibial nerve crush (DEX-DEN-R), all animals showed good plantar flexion of the right foot to stimulation, but no evidence of reinnervation on the DEX-DEN-IRR side. Statistical analysis of muscle wet weight was performed by the ANOVA test. P < 0.05 was considered significant. Half of each soleus muscle split in the middle along the longitudinal axis was removed in an isometric clamp, fixed in 2% glutaraldehyde in cacodylate buffer, weighed, and processed for resin embedding. T h e other half was weighed, then frozen in isopentane chilled with liquid nitrogen for light microscopy
and biochemistry. From the midpoint of the muscles, 8 km thick, transverse cryostat sections were reacted for myosin ATPase activity after preincubation at pH 9.4, 4.6, and 4.2. For immunocytochemistry, two monoclonal antibodies (Ab) to epitopes in the S1 (MF18) and LMM (MF20) regions of MHC were used4 (generous gift from D. Fischman, MD, Cornell University). Four-micrometer thick, transverse cryostat sections were fixed in acetone (1 minute), incubated in the undiluted primary Ab (30 minutes) followed by a biotinylated goat antimouse IgG Ab (30 minutes), which was then visualized by the streptavidin- peroxidase technique. On control sections, incubation of the primary Ab was substituted by incubation in 2% normal goat serum. Transverse and longitudinal semithin resin sections, stained with paraphenylene diamine, were examined by phase microscopy. Ultrathin sections, stained with uranyl acetate and lead citrate, were examined with an electron microscope.
''
From the animals killed 1, 2, 3, and 6 weeks after crush denervation, and from the control animals, an aliquot of frozen muscle from the right soleus was used for biochemistry. For one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) and Western blot analysis, 10 mg of frozen muscle from each of 3 or more animals of the same group were taken and pooled. Samples were homogenized and rotein was determined as previously described.' Samples were diluted with SDS sample buffer to obtain a final protein concentration of 0.5 to 1.5 mg/L. SDS-PAGE was performed according to the method of Laemmli,15 using a 12% acrylamide separating gel, loading
Biochemical Techniques.
Microscopic Techniques.
Renewal of Thick Myofilaments
1
2
3
5
4
7
Weeks After Denervation
FIGURE 1. Mean wet weight with standard deviations of soleus muscles at weekly intervals after denervation. Dexamethasone was given for the first 7 days to all animals except controls.
MUSCLE & NERVE
November 1992
1291
1.5 pg of protein per sample on a miniprotein cell (Biorad Lab, Mississauga, Ontario, Canada). Molecular weight markers were run on the same gel. The gel was stained with Coomassie blue and scanned at 550 nm using a Zeiss spectrophotometer (Model PMQII, Carl Zeiss Ltd., Oberkochen, Germany). The peak areas of actin and myosin heavy chain (MHC) bands were measured using a MOP3 image analyzer (Carl Zeiss Ltd.) and the MHC/actin ratio was calculated by taking the sum of the number of pixels times their intensity from 1 to 256. For Western blot analysis of MHC immunoreactivity, a replicate gel was electroblotted for 16 hours at 25 V in a minitransblot apparatus (Biorad) in transfer buffer (tris/glycine/20% MeOH). The blot was blocked for 45 minutes at 37°C with 10% skim milk in tris-buffered saline with Tween." The MF20 monoclonal Ab, which reacts with all MHC isoforms, was then applied at a dilution of 1:7 for 1 hour at room temperature. An antimouse IgG, alkaline-phosphatase-conjugated secondary Ab was used for visualization. RESULTS
The changes in wet weight of soleus muscles in the different experimental conditions and at different time intervals are shown in Figure 1. On the side of nerve excision (DEXDEN-IRR), the average wet weight was significantly and progressively reduced, compared with Muscle Weight.
intact controls, throughout the experiment, reaching a minimum plateau by 5 weeks. On the side of nerve crush (DEX-DEN-R), the average wet weight was reduced to approximately the same extent, at both 1 and 2 weeks. At 3, 4, and 5 weeks, the average weight, though progressively increasing, was still significantly lower than that of matched intact controls. By 7 weeks, this difference was no longer significant. The solei studied 1 week after denervation showed variability of staining for myosin ATPase at all pHs. Many fibers had surprisingly well-preserved activity, while others, regardless of fiber type, had a decreased level (Fig. 2A and B). There was conspicuous variability of staining on transverse view within single fibers. N o difference was noted between the two modalities of denervation. In DEX-DEN-R muscles studied after 3 or more weeks, the ATPase activity was normal. The MF18 Ab gave preferential staining of type I1 extra- and intrafusal fibers in control muscles, whereas MF20 gave an equal staining of all fiber types (Figs. 3A and 4A). With both Abs, the staining had a clear myofibrillar pattern. There was irregular reduction in the intensity of staining at 1 week after denervation in many muscle fibers compared with controls (Figs. 3 R and 4B). Semithin and ultrathin sections at 7 and 14 days were similar in the muscles with either Microscopic Studies.
B
A
FIGURE 2. Myosin ATPase reactions, preincubation pH 9.4. Compared with control (A), l-week DEX-DEN muscle ( 8 )shows a moderate general reduction in activity with marked focal or diffuse loss in some fibers (examples marked by arrows) (x350).
1292
Renewal of Thick Myofilaments
MUSCLE & NERVE
November 1992
A
6
FIGURE 3. Monoclonal antibody to MHC (MF18) which preferentially stains type II fibers. Compared with control (A), 1-week DEX-DEN muscle (B),shows marked focal reduction in the staining of many type II fibers (x300).
crushed or cut nerves. Varying degrees of loss of A-band density affected 40%-100ojO of fibers on individual semithin sections. The degree of loss at 14 days did not exceed that seen at 7 days, although the size of muscle fibers was smaller. The mitochondrial pattern became profoundly disturbed. On transverse sections, the normal gridlike mitochondrial densities localized to the I band
were replaced by diffusely scattered punctate densities. The rounded mitochondria normally seen aggregated beneath the sarcolemma of type I fibers were largely absent at 7 days and almost totally absent at 14 days. On longitudinal sections, mitochondria appeared elongated in the long axis of the fibers and frequently had migrated into the A band.
B
A
FIGURE 4. Antibody to MHC (MF20) which stains myofibrils of all fibers uniformly in control (A). One-week DEX-DEN muscle (6) shows less intense stain in several fibers (~300).
Renewal of Thick Myofilaments
MUSCLE & NERVE
November 1992
1293
B
A
FIGURE 5. Three weeks postdenervation. In muscle with nerve cut (A), a central fiber shows absence of A-band density. Z discs are out of register. The other fibers show mild A-band pallor and, in the fiber on the bottom, mitochondria have migrated into the A band from their usual para-Z-disc position. (B) In the reinnervated muscle, the density of A bands is normal except where sarcomeres are obliterated by Z-disc streaming, which covers a large zone, 2,in the fiber below, and is multifocal in the fiber above. Subsarcolemmal myofibril-freezone contains large dark mitochondria (arrow). Resin sections, paraphenylene diamine, phase optics ( ~ 7 5 0 ) .
After 14 days in DEX-DEN-IRR muscles, the packing density of mitochondria, as seen on transverse sections, lessened despite further shrinkage of cell girth. A band loss persisted (Fig. 5A). Marked loss of thick myofilaments was found by electron microscopy in many muscle fibers (Fig. 6A). Loss of myofilaments did not appear to exceed loss of cell girth. In fibers with extensive A-band loss, the alignment of Z discs became disturbed, but even at 7 weeks there did not appear to be any foci of total myofibrillar breakdown. Z-disc streaming was almost totally absent from DEX-DEN-IRR muscles. In DEX-DEN-R solei at 21 days, there was persistent A-band loss only in a few fibers of small size, but the majority of fibers showed either normal A-band density or extensive areas of Z-disc streaming (Figs. 5B, 6B, and 7). Many of these fibers had prominent subsarcolemmal areas that lacked myofibrils but contained large mitochondria. Such areas were never seen in fibers with A-band loss, whether focal or diffuse. At further time intervals, as the fibers regained normal size, these areas tended to be less and less prominent. Large areas of Z-disc streaming persisted; by 7 weeks they were almost the only abnormality. Foci of complete myofibrillar breakdown, signaled by lack of Z-disc density were rare and seen in only 2 specimens at 4 weeks.
1294
Renewal of Thick Myofilarnents
Biochemical Studies. One-dimensional SDSPAGE showed substantial preferential depletion of MHC relative to total protein in DEX-DEN-R muscles studied 1 week after crush denervation (Fig. 8). Because no consistent depletion of actin was evident in the electrophoretic gels, we chose MHC/actin ratio as a relative measure for MHC depletion. MHC/actin ratio was 1.5 to 1 in control solei, and it dropped to 0.78 to 1 at 1 week postdenervation (Table 1). In reinnervated muscles, by 3 weeks postdenervation, it regained a normal value (1.6 to l), which persisted at 6 weeks (1.62 to 1). No bands of abnormal molecular weight were observed in the denervated or reinnervated muscles. Three bands between MHC and glycogen phosphorylase appeared significantly paler in the 2-week DEX-DEN muscle than in the control. These were in the approximate expected location of myomesin (185 kd),6,25 M protein (165 kd), and C protein (145 kd),'8,25 but exact identification of these bands is not possible without appropriate antibodies. Western blot analysis showed that MF20 Ab immunoreactivity was localized mainly to the band corresponding to MHC in all the experimental conditions tested (not illustrated). A titer for the Ah (1:7) was chosen to allow detection of small amounts of MHC proteolytic fragments in the DEX-DEN muscles. However, such subfragments
MUSCLE & NERVE
November 1992
A
B
FIGURE 6. Three weeks postdenervation. (A) Portions of three fibers whose relative pallor indicates thick filament loss in DEX-DENIRR. Mitochondria form small random clusters. (B)Reinnervated fibers have normal myofibrils and subsarcolemmal areas with very large mitochondria, ribosomes, and glycogen. DEX-DEN-R. [Both (A) and (6)~6750.1
were not found, though overloading of the blot resulted in some diffusion of the staining to adjacent lanes and reactivity of several small bands between 80 and 100 kd in all the lanes.
DISCUSSION
In the DEX-DEN model, the combination of denervation and high dose dexamethasone for 7 days induced the following effects: (1) a reduction of the muscle wet weight by more than one third; (2) a marked reduction (-44%) of the MHC/actin ratio, revealed by SDS-PAGE; and (3) a severe preferential depletion of thick myofilaments revealed by EM. These findings are similar to those reported by Rouleau et al.,lg except that severe depletion of thick myofilaments was consistently achieved within 7 days by using higher doses of dexamethasone. The temporary denervation produced by nerve crush proved as effective in producing these changes as nerve excision. Nerve crush was always
Renewal of Thick Myofilaments
followed by reinnervation within 2 weeks. Muscle reinnervation and the withdrawal of dexamethasone resulted in rapid reconstitution of normal sarcomere pattern in all animals. By 3 weeks postdenervation, most myofibers contained normally dense A bands. Evidence that these fibers had previously been denervated was the presence of elongated mitochondria protruding into the A bands. The myosin-actin ratio returned to normal after 3 weeks, although the wet weight of the muscle was not normal until 7 weeks. Withdrawal of dexamethasone was not sufficient, by itself, for the reconstitution of thick myofilaments in permanently denervated solei. Reinnervation was an essential condition for reversibility. In most accounts of denervation, thick and thin filaments are lost together. 1922 The exception to this is from reports by Jakubiec-Puka et a1.8*Y The actin-myosin ratio dropped from 1.78 to 0.80 after 27 days. Reinnervation by 20 days after sciatic nerve crush in their experiments resulted in a relative increase in myosin filaments in all fibers,
MUSCLE 8, NERVE
November 1992
1295
FIGURE 7 . Higher magnification of Figure 6B. Note the normal myofibrils and the large myofibril-free space which contains large mitochondria, glycogen, and ribosomes ( ~ 5 0 , 0 0 0 ) .
and a return toward the regular hexagonal arrangement of thin filaments around each thick one. Myofibrils became normal in appearance in the center of fibers, while around the periphery they were small and irregular. T h e actin-myosin ratio returned to normal by 45 days. These findings after simple denervation were not confirmed by Rouleau et al.") Breakdown of actin in dener-
Table 1. Computer image analysis of gels of control and DEX-DEN-R muscles.
Control 1 week 2 weeks 3 weeks 6 weeks
Myosin
Actin
MIA
21 1494 141a25 236325 21 8757 240293
140887 180831 163001 13691 1 146583
1.50 0.78 1.45 1.60 1.64
Myosin and actin bands were circumscribed and a histogram taken of each band Data are composed of the sum of the number of pixels times their intensity on a scale of 1 to 256 in columns
1296
Renewal of Thick Myofilaments
vated muscle appears to occur by a nonlysosomal, non- Ca++-dependent,ATP-requiring system.5 Our histochemical and microscopic immunocytochemical findings were somewhat unexpected. On the basis of the EM and biochemical results, a major loss of microscopic myosin ATPase reactivity and of MHC immunoreactivity would be expected by 7 days in the majority of muscle fibers. The retention to a considerable degree of these activities could be explained if the disappearance of thick myofilaments were due to their simple disassembly into myosin monomers, which remained as a soluble cytoplasmic pool retaining ATPase activity. Myosin molecules can be disassembled in vitro from isolated thick myofilaments or from intact cells by exposure to high ionic strength solutions.' Maw and Rowe have succeeded in first solubilizing myosin and subsequently reconstructing thick myofilaments in intact cells by returning the bathing salt concentration to normal while preventing egress of the solubilized myosin molecules.16 In our model, it is
MUSCLE & NERVE
November 1992
- 92.5 - 69
- 46 1
2
3
4
5
FIGURE 8. One-dimensional SDS-PAGE stained with Coomassie blue. Lane 1: Control. Lane 2: DEX-DEN-R at 1 week with no reinnervation as yet. Lane 3: DEX-DEN-R at 2 weeks after crush with reinnervation. Lane 4: DEX-DEN-R at 3 weeks after crush with reinnewation. Lane 5: DEX-DEN-R at 6 weeks after crush with reinnervation. Molecular weights are indicated on right. Arrow indicates MHC which shows marked diminution in lane 2 and gradual recovery in lanes 3, 4, and 5. Arrowheads indicate other bands which become less dense after DEXDEN and may represent myomesin, M protein, and C protein.
noreactivity. We did not, however, detect partial proteolytic fragments by Western blot analysis. On the basis of our results, we suggest that, in the DEX-DEN model, limited proteolysis of MHC (possibly at the hinge region of the molecule) occurs first. The truncated myosin monomers then disaggregate and subsequently undergo total catabolism. The amount of remaining disaggregated, but not yet fully catabolized, MHC residues is probably sufficient to account for the microscopic partial ATPase activity and MHC immunoreactivity, in view of the degree of shrinkage occurring in the muscle fibers. Proteolysis is known to be increased in denervated m ~ s c l e ~ ~ and ” ~in* ~ steroid-treated The preferential susceptibility of MHC in the DEX-DEN model remains unexplained. The thick myofilament depletion is rapidly and fully reversible upon reinnervation of the muscle and discontinuation of the drug. For this reconstitution, de novo synthesis of MHC is probably necessary. Human disease analogous to our experiments has been reported in patients given high-dose corticosteroids while receiving pharmacologic neuromuscular blockade to facilitate artificial respiration, usually for status a s t h m a t i c u ~ .T~h*e~duration ~ of‘ neuromuscular blockade in 2 1 patients varied from 6 to 40 days. After cessation they remained quadriparetic. Recovery was generally complete over weeks to months. Biopsies in at least 3 patients showed extensive selective loss of thick myofilaments. 3 2 3
REFERENCES
possible that corticosteroids would raise the intracellular ionic concentration and also possibly that this effect would be magnified by the cell shrinkage that accompanies denervation,22 but we have no direct evidence for this. It seems unlikely that the ionic concentration needed to disassemble thick filaments into myosin monomers could be reached in vivo. If thick myofilaments had only been disassembled into individual myosin monomers, which were still present, this would account for the microscopic myosin ATPase activity and MHC immunoreactivity persisting at 1 week, but it would not be consistent with the degree of loss of MHC seen on SDS-PAGE. On the other hand, limited proteolysis of MHC would account for the SDSPAGE findings and be consistent with the persistence of myosin ATPase activity and MHC immu-
Renewal of Thick Myofilaments
1. Braunstein PW Jr. Girolami V: Experimental corticosteroid myopathy. Acta Neuropathol (Berl) 1981;55:167- 172. 2. Clarke AF, Vignos PJ Jr: Experimental corticosteroid myopathy: effect on myofibrillar ATPase activity and protein degradation. Mwcle Nerve 1979;2:205-273. 3. Danon MJ, Carpenter S: Myopathy with thick filament (myosin) loss following prolonged paralysis with vecuronium during steroid treatment. Muscle Nerve 1991; 14: 1131-1139. 4. Fischman DA, Masaki T: Immunochemical analysis of myosin with monoclonal antibodies, in Anderson W, Sadler W (eds): Perspectives in Dfferentiation and Hypertrophy. Amsterdam, Elsevier, 1980, pp 279-291. 5. Furono A, Goodman MN, Goldbert HL. Role of different systems in the degradation of muscle proteins during denervation atrophy. J Biol Chem 1990;265:8550-8557. 6. Grove BK, Kurer V, Lehner C, Doetschman TC, Perriard J-C, Eppenberger HM: A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies. J Cell Biol 1984;98:518-524. 7. Harrington WF, Rodgers ME: Myosin. Ann Rev Biochrm 1984;53:35-73. 8 . Jakubiec-Puka A, Kulesza-Lipka D, Krajewski K: The con-
MUSCLE & NERVE
November 1992
1297
tractile apparatus of striated muscle in the course of atrophy and regeneration: I. Myosin and actin filaments in the denervated rat soleus. Cell ?'issue Res 1981;220:65 1-663. 9. Jakubiec-Puka A, Kulesza-Lipka D, Kordowska J : The contractile apparatus of striated niusclc in the course of atrophy and regeneration: 11. Myosin and actin lilaments in mature rat .solcus muscle regenerating after reinnervation. Cell Ti.rsue Res 1982;227:641-650. 10. Jirmanova I, Soukup T , Zelena J : T h e pathomorphology of developing skeletal muscles of rabbits treated with glucocorticoids. Virchow's Arch (Cell Pathol) 1982;38: 323-335. 11. Johnson DA, Gautsch JW, Sportsman JR, Elder JH: Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose. Gene Anal Techniq 1984;1:3-8. 12. Karpati G, Pouliot Y, Carpenter S: Expression of immunoreactive major histocompatibility complex products in human skeletal muscles. A n n Neurol 1988;23:64-72. 13. Kelly FJ, Goldspink DF: T h e differing responses of four muscle types to dexamethasone treatment in the rat. Bzochen J 1982;208:147-151. 14. Kelly FJ, McGrath JA, Goldspink DF, Cullen MJ: A morphologicalbiochemical study on the actions of corticosteroids on rat skeletal muscle. Muscle Nerve 1986;9:1- 10. 15. Laemmli VK: Cleavage and structural proteins during the assembly of the head of bacterophage T4. Nature 1970;227:680-685. 16. Maw MC, Rowe HJ: T h e reconstitution of myosin filaments in rabbit psoas muscle from solubilized myosin. J Muscle Re.( Cell Motil 1986;7:97- 109. 17. Metafora S, Felsani A, Cotrufo K, Tajana GF, Dilorio G , Del Rio A, DePrisco PP: Neural control of gene expression in skeletal muscle fibers: the nature of the lesion in the
1298
Renewal of Thick Myofilarnents
muscular protein-syrithetizing machinery following denervation. Proc R SOCLond ( B ) 1980;209:239-273. 18. Obinata T, Maruyama K, Sugita H, Kohoma K, Ebashi S: Dynamic aspects of structural proteins in vertebrate skeletal muscle. M U X ~Nerve P 1%1;4:456-488. 19. Rouleau G , Karpati G, Carpenter S, Soza M, Prescott S, Holland P: Glucocorticoid excess induces preferential depletion of myosin in denervated skeletal muscle fibers. Muscle Nerve 1987;10:428-438. 20. Shoji S, Pennington RJT: T h e effect of cortisone on protein breakdown and synthesis in rat skeletal muscle. Mol Cell Enducrinol 1977;6: 159- 169. 21. Stern LZ, Fagan JM: 'I'he endocrine myopathies, in Vinken DJ, Bruyn GW (eds): Diseases of Muscles, Handbook of Clinical Neurology. Amsterdam, North Holland, 1979, vol 4, pp 235-258. 22. Stonnington, M M , Engel AG: Normal and denervated muscle: a morphometric study of fine structure. Neurology (Minneap) 197 3 ;23 :7 14- 724. 23. Waclawik AJ, Sufit RL, Beinlich BR, Schutta HS: Acute myopathy with selective degradation of myosin filaments following status asthmaticus. J Neurol Scz 1990;98 (suppl):470. 24. Walsh G, DeVivo D, Olson W: Histochemical and ultrastructural changes in rat muscle. Occurrence following adrenal corticotropic hormone, glucocorticoids and starvation. Arch Neurol 1972;24:83-93. 25. Wang K: Sarcomere-associated cytoskeletal lattices in striated muscle, in Shag JW (ed): Cell and Musrle Motility. New York, Plenum, 1985, vol 6, p p 315-369. 26. Weinstock Im, Iodice AA: Acid hydrolase activity in muscular dystrophy and denervation atrophy, in Dingle JT, Fell FB (eds): Lysosomes in Biology and Pathology. New York, Elsevier, 1969, p p 450-468.
MUSCLE & NERVE
November 1992
