DEVELOPMENTAL
143,422-426 (1991)
BIOLOGY
Distribution
of Myosin mRNA during Development of Skeletal Muscle Fibers DAVID J.DIx~
and Regeneration
ANDBRENDARUSSELLEISENBERG'
Department of Physiology und Biophysics, M/C 901, Uni~lersity of Illinois at Chicago, P.O. Box 6998, Chicugo, Illinois
606X0
Accepted Novembw L, 1990 distribution among subcellular compartments of anterior tibialis muscles in rabbit is monitored by in A high density of mRNA was widely distributed throughout myotubes from 29-day fetal muscle and from regenerating adult muscle. All cytoplasmic spaces contained mRNA except where scattered myofibrils and centrally located nuclei were found. In fibers from 22-week-old rabbits, myosin mRNA was concentrated under the sarcolemma and excluded from the consolidated myofibrils and peripheral nuclei. The dispersal of mRNA through the cytoplasm in myotubes suggests that translation of myosin is widespread and that rapid myofibril assembly can occur throughout the fiber. o 1931 Academic press, IX Myosin
mRNA
situ hybridization.
INTRODUCTION
MATERIALS
The distribution of intracellular mRNAs could be a mechanism by which individual cells regulate growth and repair of cytoplasmic domains. Examples of cytoplasmic domains of mRNA have been reported in cultured muscle cells (Berman et al., 1990; Lawrence and Singer, 1986) and in whole muscle (Fontaine and Changeux, 1989; Merlie and Sanes, 1985). Specific mRNA domains could be especially important in the long, multinucleated fibers of skeletal muscle. Our laboratory has used in situ hybridization (ISH) to locate high densities of myosin heavy chain (MHC) mRNA in subsarcolemmal region of adult skeletal muscle (Dix and Eisenberg, 1988). The normal mRNA pattern is altered in the mid-region and at the myotendinous junctions of fibers recovering from injury caused by stretch (Dix and Eisenberg, 1990, 1991). Co-incident with increased mRNA accumulation, we found evidence of myofibrillogenesis both in focally damaged mid-regions and at the tips of stretched fibers. In this report, we examine the distribution of MHC mRNA during developmental growth of TA fibers of rabbit and compare this with myotubes in regenerating adult muscles. We find that the distribution of mRNA among compartments changes in response to growth or repair processes. Three possible mechanisms for regulation of mRNA into cellular compartments are tested by our studies with ISH and immunocytochemistry: (1) specific association with cytoskeletal proteins, (2) incorporation of newly translated myosin into the A-band (cotranslational assembly), or (3) simple exclusion from the myofibrillar lattice. i Present address: Department of Biochemistry, State University, Raleigh, NC 27695-7622. *To whom correspondence should be addressed.
0012-1606/91$3.00 Copyright All rights
C; 1991 hy Academic Press, Inc. of reproduction in any form reserved.
North
Carolina
422
AND
METHODS
Animals and tissue. Governmental and institutional guidelines for animal care and use were followed at all times. A total of eight 29-day fetal rabbits (New Zealand White) from two litters provided TA tissue. Four 2week-old neonates and stretched muscles from four adult rabbits also provided TA muscles. Stretching of the TA was accomplished by casting the lower hind limb in full plantar flexion for 4 days. Blocks of tissue for ISH and immunofluorescence were flash-frozen in isopentane cooled with liquid nitrogen and then stored at -80°C in isopentane until use. In situ hybridization. Hybridizations were performed with pBMHC-1 as previously described (Dix and Eisenberg, 1988). Biotinylated RNA probes were transcribed from this 1107-bp fragment from the terminal region of a-cardiac MHC cDNA from rabbit, the sequence of which is 97% similar to the p-MHC (Sinha et al., 1984). Complementary anti-sense RNA probes were used experimentally; sense RNA probes were used as negative controls to rule out nonspecific RNA binding, endogenous biotin, or alkaline phosphatase activity in the tissue. Sections were hybridized at 1 ng/pl final probe concentration for 3 hr at 57°C. RNase treatment and washing steps were followed by enzymatic detection of the biotin label with streptavidin-alkaline phosphatase. Sections were viewed and photographed with a Nikon Microphot-FXA. The methods for light microscopic observation retain mRNA and preserve tissue integrity (Dix and Eisenberg, 1990, 1991), and localization has been confirmed by in situ hybridization observed with the electron microscope with different processing and detection methods (Eisenberg et al, 1991). Immunojluorescence. Immunofluorescent detection of slow MHC protein was accomplished with the monoclonal antibody HPM-7 at a 1:2000 dilution (Kennedy et al.,
BRIEF NOTES
423
FIG. 1. Myotubes seen in tibialis anterior muscle from 2%day-old fetal rabbit. (A-C) Intracellular distribution of slow myosin mRNA hyhridized it, sit/r with a biotinylated RNA probe and detected by streptavidinalkaline phosphatase to give a colored reaction product. (A) Transverse cryosection; the reaction product indicates a patchy distribution of slow myosin mRNA (arrowheads) throughout the primary myotuhr. Light areas which exclude mRNA are myofibrils (arrows) as confirmed by phase microscopy. A secondary fiber is marked with an asterisk. (8) Another primary myotube filled with mRNA (arrowheads) and with some areas devoid of myofibrils (arrows). The location of a nucleus seen in an adjacent serial section (not shown) is marked with a double-headed arrow. (C) Longitudinal cryosection of myotubr filled with mRNA (arrowheads). (D) Hcmatoxylin and eosin-stained transverse section showing peripheral nuclei (double-headed arrow). A small central area lacks a myofibril (arrowhead). Slow myosin detected by immunolluorcscencc in transverse (E) and longitudinal (F) sections shows patchy- myofibril distribution (arrows) in these primary myotuhes. Bar = 10 pm.
1986; a kind gift of Dr. R. Zak of the University of Chicago). Slow specificity of HPM-7 in rabbit was confirmed by comparison with ATPase-stained serial sections. y-A&in polyclonal antisera (the kind gift of Dr. J. Bulinski, Columbia University) was used at 150 dilution. Monoclonal antibodies against P-tubulin, desmin, vimentin, and vinculin (Sigma Chemical Co., St. Louis, MO) were used at 150 dilutions. RESIJLTS
Myofubes fi-orn ,‘9-day fetul TA. ISH of slow MHC mRNA and immunofluorescent detection of the slow MHC protein both identified the large primary myotubes in 29-day embryonic TA. Primary myotubes were surrounded by clusters of smaller secondary myotubes which did not express slow isomyosin (Fig. 1). Low-
magnification surveys of TA cross sections revealed a decreasing gradient of primary myotubes from the deep to the superficial region of the muscle (data not shown) reminiscent of the distribution of slow fibers in adult TA. The intracellular distribution of MHC mRNA in fetal myotubes was examined by ISH at higher resolution with light microscopy. Cross sections of fetal myotubes had patchy distributions of MHC mRNA between the myofibrillar masses (Fig. 1A). Zones around peripheral nuclei were densely stained for MHC mRNA (Fig. 1B). This perinuclear accumulation of mRNA extended beyond the possible dimensions of the nuclei as seen in longitudinal sections (Fig. 1C). Many fibers had densely stained central regions even though no central nuclei were observed nearby. Although the majority of nuclei were already in the subsarcolemmal region it was quite
424
DEVELOPMENTALBIOLOGY
distribution of slow mgosin mRNA in FIG. 2. (A and 8) Intracellular several fibers from TA in Z-week-old rabbits. Transverse cryosection, hybridized slow myosin mRNA is most concentrated at the periphery of fiber, very similarly to adult fibers (arrowheads). A small central area filled with mRNA is present in some fibers (arrowhead in A). (Cl Hematoxylin and eosin stain of transverse section shows perpheral nuclei (double-headed arrow) and dense mgolibrils. (Dl Immunofluorescent detection of slow mgosin shows dense distribution (arrow) with some small areas devoid of myolibrils (arrowhead). Bar = 10 lrn.
VOLUME 143.1991
ing myotubes also are found among existing fibers (Figs. 3B and 3D). MHC mRNA in all myotubes has a less-ordered intracellular distribution than was seen in healthy adult fibers. Myotubes have large dark patches of stain throughout their cross sections (Figs. 3A and 3B). Electron micrographs of these myotubes showed that the myofibrils were loosely packed, leaving much cytoplasmic space between them (our data not shown; Kelly, 1983). Immunofluorescence with the anti-slow MHC antibody resulted in a strong positive signal for over 95% of the myotubes in stretched muscles and only 5% were not positive for our slow probe and antibody (Figs. 3C and 3D). The patchiness in the fluorescent signal throughout the myotubes is due to the sparsity of myofibrils. The large central void corresponds to the nucleus and its associated cytoplasmic area. Immunofluorescence with antibodies for cytoskeletal proteins showed that y-actin microfilaments were concentrated under the sarcolemmae in developing and regenerating myotubes and adult fibers (our data not shown; Otey ef (xl., 1988). Desmin intermediate filaments were also highly concentrated under the sarcolemmae (our data not shown; Craig and Pardo, 1983). Myofibrils were outlined by desmin, but desmin was not dispersed throughout the cytoplasmic zone free of myofibrils. Yet, it is these myofibril-free zones that contain the highest mRNA accumulations. Thus, MHC mRNA accumulation was found at sites devoid of desmin and y-actin, and this result was also true for tubulin, vimentin, and vinculin (our data not shown). DISCUSSION
common to see central zones devoid of myofibrillar proteins (Figs. lD-1F). Fibers from T-week-old TA. More mature muscle fibers had more densely packed myofibrils and less diffuse mRNA distributions than fetal myotubes. Hybridizations in cross sections from these muscles exhibit reticular networks of MHC mRNA between the myofibrils (Figs. 2A and 2B). MHC mRNA is most highly segregated to the subsarcolemmal region of the fibers. Occasionally a vestige of the central concentration of mRNA is present. Hematoxylin and eosin staining confirms that the nuclei have all migrated to the subsarcolemmal region (Fig. 2C). Immunofluorescence of slow MHC in fetal fibers shows the myofibrils fill almost the entire cross-sectional area (Fig. 2D). Myotubes from regeneruting adult TA. In adult stretched muscle, the space left within the basement membrane of a degenerating fiber was often filled by myotubes (Figs. 3A and 3C) in the hyperplasia process of making new fibers (Kennedy et ul., 1988). Regenerat-
The intracellular distribution of MHC mRNA changed during development and growth of rabbit skeletal fibers. In fetal myotubes the distribution of MHC mRNA was condensed in large patches. Density of MHC mRNA staining was highest where the myofibrillar content was lowest. In 2-week-old rabbits, whose myofibrils occupy a greater fraction of the fiber, mRNA distribution more closely approximated that of adult skeletal fibers (Dix and Eisenberg, 1988). Thus, MHC mRNA was displaced to the subsarcolemmal and intermyofibrillar spaces in these more mature fibers. Possible mechanisms for the separation of MHC mRNA into compartments could be due to association of the mRNA with cytoskeletal elements, to cotranslational assembly, or to spatial exclusion of the mRNAs by myofibrils. Other laboratories have reported evidence of the association of muscle mRNAs with both the cytoskeleton and the myofibrils (Isaacs and Fulton, 1987; Praminik et al., 1986; Silva et al., 1989; Singer ef al., 1989). However, immunofluorescence on developing myotubes did not confirm any correlation between dis-
BRIEF
NOTES
425
FIG. 3. Myotubes from adult rabbit tibialis anterior injured by stretching and nestled among some normal adult fibers (asterisks). Transverse cryosections are hybridized with slow myosin mRNA (A and B) and are detected for slow myosin by immunofluorescence (C and D). Area A is from the same region as area C, and area B is from the same region as area D. Some intervening sections were used for controls so the fiber profiles do not match exactly. (A and C) Several myotubes fill the surviving basement membrane from a degenerating adult fiber. These myotubes have a patchy distribution of slow myosin mRNA (arrowheads) and myofibrils (arrows). (B and D) Myotubes also form in the endomysial connective tissue between healthy fibers, again with a patchy distribution of mRNA (arrowheads) and myofibrils (arrows). Bar = 10 pm.
tribution of MHC mRNA and several cytoskeletal proteins (desmin, y-a&in, tubulin, vimentin, and vinculin). Association with the thick filaments would result in Aband striations. Since these were not seen, we conclude that mechanisms of direct translational regulation by assembly processes are unlikely. We found mRNA everywhere except in the myofibrils and nuclei (Dix and Eisenberg, 1988; Dix and Eisenberg, 1990, 1991). There is always a possibility that some technical artifact during the many steps of ISH may result in alteration of the integrity of the structure or of mRNA redistribution. But electron microscopic ISH using different processing and detection methods in cardiac myocytes has also shown MHC mRNA in the intermyofibrillar spaces with exclusion from the filament lattice itself
(Eisenberg et al., 1991). In all situations examined so far in our laboratory, free diffusion and myofibrillar exclusion provide the simplest and most reasonable explanation of MHC mRNA distribution. Stretching of rabbit muscle is one of the ways to stimulate satellite cells which then proliferate and fuse into myotubes (Bischoff, 1990). We found distribution of regenerating myotubes of the adult similar to that in fetal myotubes: MHC mRNA had a patchy distribution throughout the cross section. The distribution of myofibrils was also patchy in these regenerating fibers but in the contrasting patches. Slow MHC is expressed by primary myotubes in fetal et al., rabbit TA as has been reported for rat (Narusawa 1987). Almost all regenerating myotubes in stretched
426
DEVELOPMENTAL
BIOLOGY
TA also expressed slow MHC as a recapitulation of the developmental program analogous to MHC expression in cold-injured (Cerny and Bandman, 1987) and stretched (Kennedy et al., 1988) chicken muscle. We believe the RNA probe and monoclonal antibody being used are specific for slow MHC because the pBMHC-1 RNA probe is 97% complementary to slow MHC mRNA and has proven specific for slow-oxidative fibers in adult rabbit muscles (Dix and Eisenberg, 1988). The HPM-7 antibody reacts with the same fibers stained by pBMHC-1 probe in serial sections. Thus we conclude that the pBMHC-1 probe and HPM-7 identify adult slow MHC. This does not exclude the possibility of reactivity with some developmental isoform. However, this paper discusses spatial distribution and our observation of MHC mRNA in the nonmyofibrillar spaces is independent of probe-specificity. Muscle gene products are restricted around nuclei in cultured myotubes (Ralston and Hall, 1989) and heterokaryons (Pavlath et al., 1989) and these nuclear domains may be the result of the distribution of mRNAs. Similarly, accumulations of MHC mRNA may restrict protein synthesis and assembly to a nuclear domain. Thus, in all cases of growth and repair that we have observed, MHC mRNA accumulates in cytoplasm that is sparsely packed with myofibrils. We suggest that accumulation of mRNA drives regional myosin translation and increases the rate of assembly into thick filaments of myofibrils nearby (Dix and Eisenberg, 1990,199l). Similarly, the distribution of MHC mRNA in developing and regenerating muscle fibers reported here could have important effects on myofibrillogenesis in regions of focal damage or rapid growth. This work was supported by the Muscular and NIH HL 40880 to B.R.E.
Dystrophy
Association
REFERENCES BERMAN, S. A., BURSZTAJN, S., BOWEN, B., AND GILBERT, W. (1990). Localization of an acetylcholine receptor intron to the nuclear membrane. Science 247,212-214. BISCHOFF, R. (1990). Cell cycle commitment of rat muscle satellite cells. Cell Biol. 111, 201-207. CERNY, L., AND BANDMAN, E. (1987). Expression of myosin heavy chain isoforms in regenerating myotubes of innervated and denervated chicken pectoral muscle. Dev. Biol. 119,350-362. CRAIG, S. W., AND PARDO, J. V. (1983). Gamma actin, spectrin, and intermediate filament proteins co-localize with vinculin at costameres, myofibril-to-sarcolemma attachment sites. Cell MotiL 3,449462. Drx, D. J., AND EISENBERG, B. R. (1988). In situ hybridization and
VOLUME
143, I991
immunocytochemistry in serial sections of rabbit skeletal muscle to detect myosin expression. J Histochem. Cytochem. 6,1519-1526. DIX, D. J., AND EISENBERG, B. R. (1990). Myosin mRNA accumulation and myofibrillogenesis at the myotendinous junction of stretched muscle fibers. J. Cell BioL 111, 1884-1894. DIX, D. J., AND EISENBERG, B. R. (1991). Redistribution of myosin heavy chain mRNA in the midregion of stretched muscle fibers, Cell Tissue Res. in press. EISENBERG, B. R., GOLDSPINK, P. H., AND WENDEROTH, M. P. (1991) Distribution of myosin heavy-chain mRNA in normal and hypertrophied heart. J. Mol. Cell Cardiol. in press. Fontaine, B., and Changeux, J. P. (1989). Localization of nicotinic acetylcholine receptor a-subunit transcripts during myogenesis and motor endplate development in the chick. J. Cell Biol. 108, 10251037. ISAACS, W. B., AND FULTON, A. B. (1987). Cotranslational assembly of myosin heavy chain in developing cultured skeletal muscle. Proc. Natl. Acud. Sci. USA 84, 6174-6178. KELLY, A. (1983). Emergence of specialization in skeletal muscle. In “Handbook of Physiology Section 10: Skeletal Muscle” (L. Peachey, Ed.), pp. 507-537. American Physiological Society, Bethesda, MD. KENNEDY, J. M., KAMEL, S., TAMBONE, W. E., VRBOVA, G., AND ZAK, R. (1986). The expression of myosin heavy chain isoforms in normal an dhypertrophied chicken slow muscle. J. Cell BioL 103, 977-983. KENNEDY, J. M., EISENBERG, B. R., REID, S., SWEENEY, L., AND ZAK, R. (1988). Nascent muscle fiber appearance in overloaded chicken slow-tonic muscle. Amer. J Anat. 181, 203-215. LAWRENCE, J. B., AND SINGER, R. H. (1986). Intracellular localization of messenger RNAs for cytoskeletal proteins. Cell, 45,407-415. MERLIE, J., AND SANES, J. (1985). C oncentration of acetylcholine receptor mRNA in synaptic regions of adult muscle fibres. Nature (Landon), 317,66-68. NARUSAWA, M., FITZSIMONS, R., IZUMO, S., NADAL-GINARD, B., RUBINSTEIN, N., AND KELLY, A. (1987). Slow myosin in developing rat skeletal muscle. J. Cell BioL 104.447-459. OTEY, C. A., KALNOSKI, M. H., AND BULINSKI, J. C. (1988). Immunolocalization of muscle and nonmuscle isoforms of actin in myogenic cells and adult skeletal muscle. Cell MotiL Cytoskeleton 9, 337-348. PAVLATH, G. K., RICH, K., WEBSTER, S. G., AND BLAU, H. M. (1989). Localization of muscle gene products in nuclear domains. Nature (London) 337,570-573. PRAMANIK, S. K., WALSH, R. W., AND BAG, J. (1986). Association of messenger RNA with the cytoskeletal framework in rat L6 myogenie cells. Euro. J. B&hem. 160,221-230. RALSTON, E., AND HALL, Z. W. (1989). Transfer of a protein encoded by a single nucleus to nearby nuclei in multinucleated myotubes. Science 244,1066-1069. SILVA, F. G., LAWRENCE, J. B., AND SINGER, R. H. (1989). Progress toward ultrastructural identification of individual mRNAs in thin section: Myosin heavy chain mRNA in developing myotubes. Tech. Immunocytochem. 4,147-165. SINHA, A. M., FRIEDMAN, D. J., NIGRO, J. M., JAKOVCIC, S., RABINOwrrz, M., AND UMEDA, P. K. (1984). Expression of rabbit ventricular n-myosin heavy chain messenger RNA sequences in atria1 muscle. J. BioL Chem. 259,6674-6680. SINGER, R. H., LANGEVIN, G. L., AND LAWRENCE, J. B. (1989). Ultrastructural visualization of cytoskeletal mNRAs and their associated proteins using double-label in situ hybridization. J. Cell BioL 108, 2343-2353.
143,422-426 (1991)
BIOLOGY
Distribution
of Myosin mRNA during Development of Skeletal Muscle Fibers DAVID J.DIx~
and Regeneration
ANDBRENDARUSSELLEISENBERG'
Department of Physiology und Biophysics, M/C 901, Uni~lersity of Illinois at Chicago, P.O. Box 6998, Chicugo, Illinois
606X0
Accepted Novembw L, 1990 distribution among subcellular compartments of anterior tibialis muscles in rabbit is monitored by in A high density of mRNA was widely distributed throughout myotubes from 29-day fetal muscle and from regenerating adult muscle. All cytoplasmic spaces contained mRNA except where scattered myofibrils and centrally located nuclei were found. In fibers from 22-week-old rabbits, myosin mRNA was concentrated under the sarcolemma and excluded from the consolidated myofibrils and peripheral nuclei. The dispersal of mRNA through the cytoplasm in myotubes suggests that translation of myosin is widespread and that rapid myofibril assembly can occur throughout the fiber. o 1931 Academic press, IX Myosin
mRNA
situ hybridization.
INTRODUCTION
MATERIALS
The distribution of intracellular mRNAs could be a mechanism by which individual cells regulate growth and repair of cytoplasmic domains. Examples of cytoplasmic domains of mRNA have been reported in cultured muscle cells (Berman et al., 1990; Lawrence and Singer, 1986) and in whole muscle (Fontaine and Changeux, 1989; Merlie and Sanes, 1985). Specific mRNA domains could be especially important in the long, multinucleated fibers of skeletal muscle. Our laboratory has used in situ hybridization (ISH) to locate high densities of myosin heavy chain (MHC) mRNA in subsarcolemmal region of adult skeletal muscle (Dix and Eisenberg, 1988). The normal mRNA pattern is altered in the mid-region and at the myotendinous junctions of fibers recovering from injury caused by stretch (Dix and Eisenberg, 1990, 1991). Co-incident with increased mRNA accumulation, we found evidence of myofibrillogenesis both in focally damaged mid-regions and at the tips of stretched fibers. In this report, we examine the distribution of MHC mRNA during developmental growth of TA fibers of rabbit and compare this with myotubes in regenerating adult muscles. We find that the distribution of mRNA among compartments changes in response to growth or repair processes. Three possible mechanisms for regulation of mRNA into cellular compartments are tested by our studies with ISH and immunocytochemistry: (1) specific association with cytoskeletal proteins, (2) incorporation of newly translated myosin into the A-band (cotranslational assembly), or (3) simple exclusion from the myofibrillar lattice. i Present address: Department of Biochemistry, State University, Raleigh, NC 27695-7622. *To whom correspondence should be addressed.
0012-1606/91$3.00 Copyright All rights
C; 1991 hy Academic Press, Inc. of reproduction in any form reserved.
North
Carolina
422
AND
METHODS
Animals and tissue. Governmental and institutional guidelines for animal care and use were followed at all times. A total of eight 29-day fetal rabbits (New Zealand White) from two litters provided TA tissue. Four 2week-old neonates and stretched muscles from four adult rabbits also provided TA muscles. Stretching of the TA was accomplished by casting the lower hind limb in full plantar flexion for 4 days. Blocks of tissue for ISH and immunofluorescence were flash-frozen in isopentane cooled with liquid nitrogen and then stored at -80°C in isopentane until use. In situ hybridization. Hybridizations were performed with pBMHC-1 as previously described (Dix and Eisenberg, 1988). Biotinylated RNA probes were transcribed from this 1107-bp fragment from the terminal region of a-cardiac MHC cDNA from rabbit, the sequence of which is 97% similar to the p-MHC (Sinha et al., 1984). Complementary anti-sense RNA probes were used experimentally; sense RNA probes were used as negative controls to rule out nonspecific RNA binding, endogenous biotin, or alkaline phosphatase activity in the tissue. Sections were hybridized at 1 ng/pl final probe concentration for 3 hr at 57°C. RNase treatment and washing steps were followed by enzymatic detection of the biotin label with streptavidin-alkaline phosphatase. Sections were viewed and photographed with a Nikon Microphot-FXA. The methods for light microscopic observation retain mRNA and preserve tissue integrity (Dix and Eisenberg, 1990, 1991), and localization has been confirmed by in situ hybridization observed with the electron microscope with different processing and detection methods (Eisenberg et al, 1991). Immunojluorescence. Immunofluorescent detection of slow MHC protein was accomplished with the monoclonal antibody HPM-7 at a 1:2000 dilution (Kennedy et al.,
BRIEF NOTES
423
FIG. 1. Myotubes seen in tibialis anterior muscle from 2%day-old fetal rabbit. (A-C) Intracellular distribution of slow myosin mRNA hyhridized it, sit/r with a biotinylated RNA probe and detected by streptavidinalkaline phosphatase to give a colored reaction product. (A) Transverse cryosection; the reaction product indicates a patchy distribution of slow myosin mRNA (arrowheads) throughout the primary myotuhr. Light areas which exclude mRNA are myofibrils (arrows) as confirmed by phase microscopy. A secondary fiber is marked with an asterisk. (8) Another primary myotube filled with mRNA (arrowheads) and with some areas devoid of myofibrils (arrows). The location of a nucleus seen in an adjacent serial section (not shown) is marked with a double-headed arrow. (C) Longitudinal cryosection of myotubr filled with mRNA (arrowheads). (D) Hcmatoxylin and eosin-stained transverse section showing peripheral nuclei (double-headed arrow). A small central area lacks a myofibril (arrowhead). Slow myosin detected by immunolluorcscencc in transverse (E) and longitudinal (F) sections shows patchy- myofibril distribution (arrows) in these primary myotuhes. Bar = 10 pm.
1986; a kind gift of Dr. R. Zak of the University of Chicago). Slow specificity of HPM-7 in rabbit was confirmed by comparison with ATPase-stained serial sections. y-A&in polyclonal antisera (the kind gift of Dr. J. Bulinski, Columbia University) was used at 150 dilution. Monoclonal antibodies against P-tubulin, desmin, vimentin, and vinculin (Sigma Chemical Co., St. Louis, MO) were used at 150 dilutions. RESIJLTS
Myofubes fi-orn ,‘9-day fetul TA. ISH of slow MHC mRNA and immunofluorescent detection of the slow MHC protein both identified the large primary myotubes in 29-day embryonic TA. Primary myotubes were surrounded by clusters of smaller secondary myotubes which did not express slow isomyosin (Fig. 1). Low-
magnification surveys of TA cross sections revealed a decreasing gradient of primary myotubes from the deep to the superficial region of the muscle (data not shown) reminiscent of the distribution of slow fibers in adult TA. The intracellular distribution of MHC mRNA in fetal myotubes was examined by ISH at higher resolution with light microscopy. Cross sections of fetal myotubes had patchy distributions of MHC mRNA between the myofibrillar masses (Fig. 1A). Zones around peripheral nuclei were densely stained for MHC mRNA (Fig. 1B). This perinuclear accumulation of mRNA extended beyond the possible dimensions of the nuclei as seen in longitudinal sections (Fig. 1C). Many fibers had densely stained central regions even though no central nuclei were observed nearby. Although the majority of nuclei were already in the subsarcolemmal region it was quite
424
DEVELOPMENTALBIOLOGY
distribution of slow mgosin mRNA in FIG. 2. (A and 8) Intracellular several fibers from TA in Z-week-old rabbits. Transverse cryosection, hybridized slow myosin mRNA is most concentrated at the periphery of fiber, very similarly to adult fibers (arrowheads). A small central area filled with mRNA is present in some fibers (arrowhead in A). (Cl Hematoxylin and eosin stain of transverse section shows perpheral nuclei (double-headed arrow) and dense mgolibrils. (Dl Immunofluorescent detection of slow mgosin shows dense distribution (arrow) with some small areas devoid of myolibrils (arrowhead). Bar = 10 lrn.
VOLUME 143.1991
ing myotubes also are found among existing fibers (Figs. 3B and 3D). MHC mRNA in all myotubes has a less-ordered intracellular distribution than was seen in healthy adult fibers. Myotubes have large dark patches of stain throughout their cross sections (Figs. 3A and 3B). Electron micrographs of these myotubes showed that the myofibrils were loosely packed, leaving much cytoplasmic space between them (our data not shown; Kelly, 1983). Immunofluorescence with the anti-slow MHC antibody resulted in a strong positive signal for over 95% of the myotubes in stretched muscles and only 5% were not positive for our slow probe and antibody (Figs. 3C and 3D). The patchiness in the fluorescent signal throughout the myotubes is due to the sparsity of myofibrils. The large central void corresponds to the nucleus and its associated cytoplasmic area. Immunofluorescence with antibodies for cytoskeletal proteins showed that y-actin microfilaments were concentrated under the sarcolemmae in developing and regenerating myotubes and adult fibers (our data not shown; Otey ef (xl., 1988). Desmin intermediate filaments were also highly concentrated under the sarcolemmae (our data not shown; Craig and Pardo, 1983). Myofibrils were outlined by desmin, but desmin was not dispersed throughout the cytoplasmic zone free of myofibrils. Yet, it is these myofibril-free zones that contain the highest mRNA accumulations. Thus, MHC mRNA accumulation was found at sites devoid of desmin and y-actin, and this result was also true for tubulin, vimentin, and vinculin (our data not shown). DISCUSSION
common to see central zones devoid of myofibrillar proteins (Figs. lD-1F). Fibers from T-week-old TA. More mature muscle fibers had more densely packed myofibrils and less diffuse mRNA distributions than fetal myotubes. Hybridizations in cross sections from these muscles exhibit reticular networks of MHC mRNA between the myofibrils (Figs. 2A and 2B). MHC mRNA is most highly segregated to the subsarcolemmal region of the fibers. Occasionally a vestige of the central concentration of mRNA is present. Hematoxylin and eosin staining confirms that the nuclei have all migrated to the subsarcolemmal region (Fig. 2C). Immunofluorescence of slow MHC in fetal fibers shows the myofibrils fill almost the entire cross-sectional area (Fig. 2D). Myotubes from regeneruting adult TA. In adult stretched muscle, the space left within the basement membrane of a degenerating fiber was often filled by myotubes (Figs. 3A and 3C) in the hyperplasia process of making new fibers (Kennedy et ul., 1988). Regenerat-
The intracellular distribution of MHC mRNA changed during development and growth of rabbit skeletal fibers. In fetal myotubes the distribution of MHC mRNA was condensed in large patches. Density of MHC mRNA staining was highest where the myofibrillar content was lowest. In 2-week-old rabbits, whose myofibrils occupy a greater fraction of the fiber, mRNA distribution more closely approximated that of adult skeletal fibers (Dix and Eisenberg, 1988). Thus, MHC mRNA was displaced to the subsarcolemmal and intermyofibrillar spaces in these more mature fibers. Possible mechanisms for the separation of MHC mRNA into compartments could be due to association of the mRNA with cytoskeletal elements, to cotranslational assembly, or to spatial exclusion of the mRNAs by myofibrils. Other laboratories have reported evidence of the association of muscle mRNAs with both the cytoskeleton and the myofibrils (Isaacs and Fulton, 1987; Praminik et al., 1986; Silva et al., 1989; Singer ef al., 1989). However, immunofluorescence on developing myotubes did not confirm any correlation between dis-
BRIEF
NOTES
425
FIG. 3. Myotubes from adult rabbit tibialis anterior injured by stretching and nestled among some normal adult fibers (asterisks). Transverse cryosections are hybridized with slow myosin mRNA (A and B) and are detected for slow myosin by immunofluorescence (C and D). Area A is from the same region as area C, and area B is from the same region as area D. Some intervening sections were used for controls so the fiber profiles do not match exactly. (A and C) Several myotubes fill the surviving basement membrane from a degenerating adult fiber. These myotubes have a patchy distribution of slow myosin mRNA (arrowheads) and myofibrils (arrows). (B and D) Myotubes also form in the endomysial connective tissue between healthy fibers, again with a patchy distribution of mRNA (arrowheads) and myofibrils (arrows). Bar = 10 pm.
tribution of MHC mRNA and several cytoskeletal proteins (desmin, y-a&in, tubulin, vimentin, and vinculin). Association with the thick filaments would result in Aband striations. Since these were not seen, we conclude that mechanisms of direct translational regulation by assembly processes are unlikely. We found mRNA everywhere except in the myofibrils and nuclei (Dix and Eisenberg, 1988; Dix and Eisenberg, 1990, 1991). There is always a possibility that some technical artifact during the many steps of ISH may result in alteration of the integrity of the structure or of mRNA redistribution. But electron microscopic ISH using different processing and detection methods in cardiac myocytes has also shown MHC mRNA in the intermyofibrillar spaces with exclusion from the filament lattice itself
(Eisenberg et al., 1991). In all situations examined so far in our laboratory, free diffusion and myofibrillar exclusion provide the simplest and most reasonable explanation of MHC mRNA distribution. Stretching of rabbit muscle is one of the ways to stimulate satellite cells which then proliferate and fuse into myotubes (Bischoff, 1990). We found distribution of regenerating myotubes of the adult similar to that in fetal myotubes: MHC mRNA had a patchy distribution throughout the cross section. The distribution of myofibrils was also patchy in these regenerating fibers but in the contrasting patches. Slow MHC is expressed by primary myotubes in fetal et al., rabbit TA as has been reported for rat (Narusawa 1987). Almost all regenerating myotubes in stretched
426
DEVELOPMENTAL
BIOLOGY
TA also expressed slow MHC as a recapitulation of the developmental program analogous to MHC expression in cold-injured (Cerny and Bandman, 1987) and stretched (Kennedy et al., 1988) chicken muscle. We believe the RNA probe and monoclonal antibody being used are specific for slow MHC because the pBMHC-1 RNA probe is 97% complementary to slow MHC mRNA and has proven specific for slow-oxidative fibers in adult rabbit muscles (Dix and Eisenberg, 1988). The HPM-7 antibody reacts with the same fibers stained by pBMHC-1 probe in serial sections. Thus we conclude that the pBMHC-1 probe and HPM-7 identify adult slow MHC. This does not exclude the possibility of reactivity with some developmental isoform. However, this paper discusses spatial distribution and our observation of MHC mRNA in the nonmyofibrillar spaces is independent of probe-specificity. Muscle gene products are restricted around nuclei in cultured myotubes (Ralston and Hall, 1989) and heterokaryons (Pavlath et al., 1989) and these nuclear domains may be the result of the distribution of mRNAs. Similarly, accumulations of MHC mRNA may restrict protein synthesis and assembly to a nuclear domain. Thus, in all cases of growth and repair that we have observed, MHC mRNA accumulates in cytoplasm that is sparsely packed with myofibrils. We suggest that accumulation of mRNA drives regional myosin translation and increases the rate of assembly into thick filaments of myofibrils nearby (Dix and Eisenberg, 1990,199l). Similarly, the distribution of MHC mRNA in developing and regenerating muscle fibers reported here could have important effects on myofibrillogenesis in regions of focal damage or rapid growth. This work was supported by the Muscular and NIH HL 40880 to B.R.E.
Dystrophy
Association
REFERENCES BERMAN, S. A., BURSZTAJN, S., BOWEN, B., AND GILBERT, W. (1990). Localization of an acetylcholine receptor intron to the nuclear membrane. Science 247,212-214. BISCHOFF, R. (1990). Cell cycle commitment of rat muscle satellite cells. Cell Biol. 111, 201-207. CERNY, L., AND BANDMAN, E. (1987). Expression of myosin heavy chain isoforms in regenerating myotubes of innervated and denervated chicken pectoral muscle. Dev. Biol. 119,350-362. CRAIG, S. W., AND PARDO, J. V. (1983). Gamma actin, spectrin, and intermediate filament proteins co-localize with vinculin at costameres, myofibril-to-sarcolemma attachment sites. Cell MotiL 3,449462. Drx, D. J., AND EISENBERG, B. R. (1988). In situ hybridization and
VOLUME
143, I991
immunocytochemistry in serial sections of rabbit skeletal muscle to detect myosin expression. J Histochem. Cytochem. 6,1519-1526. DIX, D. J., AND EISENBERG, B. R. (1990). Myosin mRNA accumulation and myofibrillogenesis at the myotendinous junction of stretched muscle fibers. J. Cell BioL 111, 1884-1894. DIX, D. J., AND EISENBERG, B. R. (1991). Redistribution of myosin heavy chain mRNA in the midregion of stretched muscle fibers, Cell Tissue Res. in press. EISENBERG, B. R., GOLDSPINK, P. H., AND WENDEROTH, M. P. (1991) Distribution of myosin heavy-chain mRNA in normal and hypertrophied heart. J. Mol. Cell Cardiol. in press. Fontaine, B., and Changeux, J. P. (1989). Localization of nicotinic acetylcholine receptor a-subunit transcripts during myogenesis and motor endplate development in the chick. J. Cell Biol. 108, 10251037. ISAACS, W. B., AND FULTON, A. B. (1987). Cotranslational assembly of myosin heavy chain in developing cultured skeletal muscle. Proc. Natl. Acud. Sci. USA 84, 6174-6178. KELLY, A. (1983). Emergence of specialization in skeletal muscle. In “Handbook of Physiology Section 10: Skeletal Muscle” (L. Peachey, Ed.), pp. 507-537. American Physiological Society, Bethesda, MD. KENNEDY, J. M., KAMEL, S., TAMBONE, W. E., VRBOVA, G., AND ZAK, R. (1986). The expression of myosin heavy chain isoforms in normal an dhypertrophied chicken slow muscle. J. Cell BioL 103, 977-983. KENNEDY, J. M., EISENBERG, B. R., REID, S., SWEENEY, L., AND ZAK, R. (1988). Nascent muscle fiber appearance in overloaded chicken slow-tonic muscle. Amer. J Anat. 181, 203-215. LAWRENCE, J. B., AND SINGER, R. H. (1986). Intracellular localization of messenger RNAs for cytoskeletal proteins. Cell, 45,407-415. MERLIE, J., AND SANES, J. (1985). C oncentration of acetylcholine receptor mRNA in synaptic regions of adult muscle fibres. Nature (Landon), 317,66-68. NARUSAWA, M., FITZSIMONS, R., IZUMO, S., NADAL-GINARD, B., RUBINSTEIN, N., AND KELLY, A. (1987). Slow myosin in developing rat skeletal muscle. J. Cell BioL 104.447-459. OTEY, C. A., KALNOSKI, M. H., AND BULINSKI, J. C. (1988). Immunolocalization of muscle and nonmuscle isoforms of actin in myogenic cells and adult skeletal muscle. Cell MotiL Cytoskeleton 9, 337-348. PAVLATH, G. K., RICH, K., WEBSTER, S. G., AND BLAU, H. M. (1989). Localization of muscle gene products in nuclear domains. Nature (London) 337,570-573. PRAMANIK, S. K., WALSH, R. W., AND BAG, J. (1986). Association of messenger RNA with the cytoskeletal framework in rat L6 myogenie cells. Euro. J. B&hem. 160,221-230. RALSTON, E., AND HALL, Z. W. (1989). Transfer of a protein encoded by a single nucleus to nearby nuclei in multinucleated myotubes. Science 244,1066-1069. SILVA, F. G., LAWRENCE, J. B., AND SINGER, R. H. (1989). Progress toward ultrastructural identification of individual mRNAs in thin section: Myosin heavy chain mRNA in developing myotubes. Tech. Immunocytochem. 4,147-165. SINHA, A. M., FRIEDMAN, D. J., NIGRO, J. M., JAKOVCIC, S., RABINOwrrz, M., AND UMEDA, P. K. (1984). Expression of rabbit ventricular n-myosin heavy chain messenger RNA sequences in atria1 muscle. J. BioL Chem. 259,6674-6680. SINGER, R. H., LANGEVIN, G. L., AND LAWRENCE, J. B. (1989). Ultrastructural visualization of cytoskeletal mNRAs and their associated proteins using double-label in situ hybridization. J. Cell BioL 108, 2343-2353.
