Fixation and immunofluorescent analysis of creatine kinase isozymes in embryonic skeletal muscle

Pectoral muscles from chicken embryos of various ages were examined with immunofluorescent and radiolabeled probes for the presence of brain-type creatine kinase (B-CK), muscle-specific creatine kinase (M-CK), muscle-specific myosin heavy chain (MHC), and cycling cells. The diffusible creatine kinase isozymes were not detectable by indirect immunofluorescence after standard histological fixation of embryonic muscle. However, a fixation procedure was devised that permitted immunodetection of the creatine kinase isozymes (particularly B-CK) in embryonic tissue from all stages of development studied. B-CK, M-CK, and MHC were all detected in post-mitotic muscle cells, but only B-CK was detected in cycling cells. Correlations between these findings and in vitro observations of a deterministic muscle lineage are discussed.     

Turnover of the creatine kinase subunits in chicken myogenic cell cultures and in fibroblasts.

The rates of degradation of creatine kinase subunits, M-CK and B-CK subunits, were measured in cultured myogenic cells and in subcultured fibroblasts. In differentiated myogenic cells, the myotubes, both M-CK and B-CK subunits are synthesized. Their rates of degradation were compared. The M-CK subunits is slightly more stable and is degraded with an average apparent half-life of 75 h, whereas that of the B-CK subunit was shorter with 63 h. The turnover properties of M-CK subunit from soluble and of myofibril-bound MM-CK homodimeric creatine kinase isoenzyme isolated from breast muscle of young chickens were identical. The apparent half-life of the B-CK subunit was also determined in subcultured fibroblasts and 5-bromo-2'-deoxyuridine-treated cells, and found to be shorter than in myotubes (46 h and 37 h respectively). Similar observations were made for myosin heavy chain, actin and total acid-precipitable material. It appears therefore that proteins are in general degraded more slowly in differentiated myogenic cells. The differences in the stability of M-CK and B-CK subunits in myotubes probably do not reflect a major regulatory mechanism of the creatine kinase isoenzyme transition.     

Regulation of creatine kinase induction in differentiating mouse myoblasts.

The regulation of creatine kinase (CK) induction during muscle differentiation was analyzed with MM14 mouse myoblasts. These cells withdraw from the cell cycle and commit to terminal differentiation when fed with mitogen-depleted medium. Myoblasts contained trace amounts of an isozyme of brain CK (designated BB-CK), but differentiation was accompanied by the induction of two other isozymes of muscle and brain CKs (designated MM-CK and MB-CK). Increased CK activity was detectable within 6 h of mitogen removal, 3 h after the first cells committed to differentiation and 6 h before fusion began. By 48 h, MM-CK activity increased more than 400-fold, MB-CK activity increased more than 150-fold, and BB-CK activity increased more than 10-fold. Antibodies prepared against purified mouse MM-CK cross-reacted with muscle and brain CKs (designated M-CK and B-CK, respectively) from a variety of species and were used to demonstrate that the increase in enzymatic activity was paralleled by an increase in the protein itself. CK antibodies were also used to aid in identifying cDNA clones to M-CK. cDNA sequences which corresponded to protein-coding regions cross-hybridized with B-CK mRNA; however, a subclone containing the 3'-nontranslated region was unique and was used to quantitate M-CK mRNA levels during myoblast differentiation. M-CK mRNA was not detectable in myoblasts, but within 5 to 6 h of mitogen withdrawal (6 to 7 h before fusion begins) it accumulated to about 30 molecules per cell. By 24 h, myotubes contained approximately 1,100 molecules per nucleus of M-CK mRNA.     

Analysis of the myogenic lineage in chick embryos

Abstract. Probabilistic and programmed lineage models for the generation of terminally differentiated skeletal muscle cells were tested in a clonal culture assay. Myogenic cells from the breast muscles of 10-day chick embryos were plated at an initial density of 250–1000 cells per 60 mm dish. Well-isolated individual cells were marked with a ring on the underside of the dishes, and clones arising from only these cells were followed. The presence of post-mitotic myoblasts in clones was assayed by peroxidase-antiperoxidase (PAP) and fluorescence immunocytochemical staining for both M-type creatine kinase (MCK) and skeletal muscle myosin heavy chain (MHC). Clones were fixed at intervals up to 76 h and were scored for the number of cells per clone and the number of MCK+ and MHC+ cells per clone. Quantitative and kinetic data were obtained indicating that post-mitotic myoblasts occurred overwhelmingly in homogeneous clones (all cells MCK⁺ and MHC⁺) which contained 2ⁿ cells ( n =0, 1, 2, 3, 4). This result does not support either probabilistic models of myogenesis or the existence of 'proliferative' mitoses at the end stages of differentiation. Rather, it indicates that myogenic precursor cells are a heterogeneous population, within which individual cells are predetermined to undergo a set number of symmetrical mitoses prior to yielding terminally differentiated progeny. These findings are strong evidence for a programmed, cell cycle-dependent lineage in the end stages of muscle differentiation.     

Intracellular targeting of isoproteins in muscle cytoarchitecture

Part of the muscle creatine kinase (MM-CK) in skeletal muscle of chicken is localized in the M-band of myofibrils, while chicken heart cells containing myofibrils and BB-CK, but not expressing MM-CK, do not show this association. The specificity of the MM-CK interaction was tested using cultured chicken heart cells as "living test tubes" by microinjection of in vitro generated MM-CK and hybrid M-CK/B-CK mRNA with SP6 RNA polymerase. The resulting translation products were detected in injected cells with isoprotein-specific antibodies. M-CK molecules and translation products of chimeric cDNA molecules containing the head half of the B-CK and the tail half of the M-CK coding regions were localized in the M-band of the myofibrils. The tail, but not the head portion of M-CK is essential for the association of M-CK with the M-band of myofibrils. We conclude that gross biochemical properties do not always coincide with a molecule's specific functions like the participation in cell cytoarchitecture which may depend on molecular targeting even within the same cellular compartment.     

Accumulation of creatine kinase mRNA during myogenesis: molecular cloning of a B-creatine kinase cDNA

Cytosolic creatine kinase isoenzymes MM, MB, and BB are assembled from M or B subunits which occur in different relative amounts in specific tissues. The accumulation of mRNAs encoding the M and B subunits was measured during myogenesis in culture. The relative concentration of the two mRNAs was determined by hybridization with a M-CK cDNA probe isolated previously and a B-CK cDNA probe, the cloning and characterization of which is reported here. The B-CK cDNA hybridizes specifically to a 1.6-kb mRNA found in brain and gizzard but not in adult skeletal muscle tissue. The M-CK cDNA hybridizes to a smaller mRNA 1.4-kb long which is specific to skeletal muscle. In culture, the B-CK mRNA is transiently induced and then declines to a low but detectable level.     

Approaching the multifaceted nature of energy metabolism: inactivation of the cytosolic creatine kinases via homologous recombination in mouse embryonic stem cells

To study the physiological role of the creatine kinase/phosphocreatine (CK/PCr) system in cells and tissues with a high and fluctuating energy demand we have concentrated on the site-directed inactivation of the B- and M-CK genes encoding the cytosolic CK protein subunits. In our approach we used homologous recombination in mouse embryonic stem (ES) cells from strain 129/Sv. Using targeting constructs based on strain 129/Sv isogenic DNA we managed to ablate the essential exons of the B-CK and M-CK genes at reasonably high frequencies. ES clones with fully disrupted B-CK and two types of M-CK gene mutations, a null (M-CK^–) and leaky (M-CK¹) mutation, were used to generate chimaeric mutant mice via injection in strain C57BL/6 derived blastocysts. Chimaeras with the B-CK null mutation have no overt abnormalities but failed to transmit the mutation to their offspring. For the M-CK^– and M-CK¹ mutations successful transmission was achieved and heterozygous and homozygous mutant mice were bred. Animals deficient in MM-CK are phenotypically normal but lack muscular burst activity. Fluxes through the CK reaction in skeletal muscle are highly impaired and fast fibres show adaptation in cellular architecture and storage of glycogen. Mice homozygous for the leaky M-CK allele, which have 3-fold reduced MM-CK activity, show normal fast fibres but CK fluxes and burst activity are still not restored to wildtype levels.     

Analysis of the myogenic lineage in chick embryos. IV. Effects of conditioned medium

Immunocytochemical analysis of small myogenic clones was used to compare the effects of fresh medium (FM) and conditioned medium (CM) on muscle differentiation. In order to compare the same population of cells, clones were initiated in FM and then switched to either new FM or to CM. Clones were fixed at 12-hour intervals up to 76 hours, then assayed for the presence of post-mitotic myoblasts by immunoperoxidase staining for muscle myosin heavy chain (MHC) or M-creatine kinase (MCK). In both media, myogenic cells occurred predominantly in homogeneous positive clones (all cells (MHC +/MCK +) which contained 2" cells. At 76 hours, the percentages of 1-, 2-, and 4-cell positive clones did not differ statistically in the two conditions; however, the percentages of 8- and 16-cell positive clones were significantly reduced in CM, and the percentages of small negative clones were concomitantly increased. We conclude from these data that CM affects myogenesis by slowing progression through a predetermined lineage rather than by changing the number of mitoses an individual cell will undergo before terminally differentiating. These results further support the idea that progress through the myogenic lineage is mediated by cell divisions.     

Multiple isoforms and an unusual cathodic isoform of creatine kinase from channel catfish (Ictalurus punctatus)

In vertebrates, the creatine kinase (CK) family consists of two cytosolic and two mitochondrial isoforms. The two cytosolic isoforms are the muscle type (M-CK) and the brain type (B-CK). Here we report multiple CK isoenzymes in the diploid channel catfish (Ictalurus punctatus) with one unusual cathodic isoform that was previously found only in pathological situations in human. The cathodic CK isoform existed only in the channel catfish stomach, ovary, and spleen, but not in any other species analyzed such as tilapia, smallmouth bass, chicken, or rat. Two genes encode the multiple forms of the channel catfish M-CK cDNAs. M-CK1 has three alleles, M-CK1.1, M-CK1.2, and M-CK1.3, while M-CK2 has just one allele as determined by analysis of 17 cDNA clones and by allele-specific PCR. M-CK1 encodes a protein of 381 amino acids and the M-CK2 cDNA encodes a protein of 380 amino acids. The two cDNAs shared an 86% identity and both have the nine diagnostic boxes for cytosolic CKs and thus are of cytosolic origin. The M-CK1 gene was isolated, sequenced, and characterized and its promoter should be useful for transgenic research for muscle-specific expression.     

MyoD and myogenin expression patterns in cultures of fetal and adult chicken myoblasts.

Isolated chicken myoblasts had previously been utilized in many studies aiming at understanding the emergence and regulation of the adult myogenic precursors (satellite cells). However, in recent years only a small number of chicken satellite cell studies have been published compared to the increasing number of studies with rodent satellite cells. In large part this is due to the lack of markers for tracing avian myogenic cells before they become terminally differentiated and express muscle-specific structural proteins. We previously demonstrated that myoblasts isolated from fetal and adult chicken muscle display distinct schedules of myosin heavy-chain isoform expression in culture. We further showed that myoblasts isolated from newly hatched and young chickens already possess the adult myoblast phenotype. In this article, we report on the use of polyclonal antibodies against the chicken myogenic regulatory factor proteins MyoD and myogenin for monitoring fetal and adult chicken myoblasts as they progress from proliferation to differentiation in culture. Fetal-type myoblasts were isolated from 11-day-old embryos and adult-type myoblasts were isolated from 3-week-old chickens. We conclude that fetal myoblasts express both MyoD and myogenin within the first day in culture and rapidly transit into the differentiated myosin-expressing state. In contrast, adult myoblasts are essentially negative for MyoD and myogenin by culture Day 1 and subsequently express first MyoD and then myogenin before expressing sarcomeric myosin. The delayed MyoD-to-myogenin transition in adult myoblasts is accompanied by a lag in the fusion into myotubes, compared to fetal myoblasts. We also report on the use of a commercial antibody against the myocyte enhancer factor 2A (MEF2A) to detect terminally differentiated chicken myoblasts by their MEF2+ nuclei. Collectively, the results support the hypothesis that fetal and adult myoblasts represent different phenotypic populations. The fetal myoblasts may already be destined for terminal differentiation at the time of their isolation, and the adult myoblasts may represent progenitors that reside in an earlier compartment of the myogenic lineage.     

Localized cyclic AMP-dependent protein kinase activity is required for myogenic cell fusion

Multinucleated myotubes are formed by fusion of mononucleated myogenic progenitor cells (myoblasts) during terminal skeletal muscle differentiation. In addition, myoblasts fuse with myotubes, but terminally differentiated myotubes have not been shown to fuse with each other. We show here that an adenylate cyclase activator, forskolin, and other reagents that elevate intracellular cyclic AMP (cAMP) levels induced cell fusion between small bipolar myotubes in vitro. Then an extra-large myotube, designated a "myosheet," was produced by both primary and established mouse myogenic cells. Myotube-to-myotube fusion always occurred between the leading edge of lamellipodia at the polar end of one myotube and the lateral plasma membrane of the other. Forskolin enhanced the formation of lamellipodia where cAMP-dependent protein kinase (PKA) was accumulated. Blocking enzymatic activity or anchoring of PKA suppressed forskolin-enhanced lamellipodium formation and prevented fusion of multinucleated myotubes. Localized PKA activity was also required for fusion of mononucleated myoblasts. The present results suggest that localized PKA plays a pivotal role in the early steps of myogenic cell fusion, such as cell-to-cell contact/recognition through lamellipodium formation. Furthermore, the localized cAMP-PKA pathway might be involved in the specification of the fusion-competent areas of the plasma membrane in lamellipodia of myogenic cells.     

Discrete subcellular localization of a cytoplasmic and a mitochondrial isozyme of creatine kinase in intestinal epithelial cells

Two isozymes of creatine kinase have been purified differentially from mitochondrial and cytoplasmic subfractions of intestinal epithelial cells. These intestinal epithelial cell creatine kinases were indistinguishable from the cytoplasmic (B-CK) and mitochondrial (Mi-CK) creatine kinase isozymes of brain when compared by SDS-PAGE, cellulose polyacetate electrophoresis, and peptide mapping. In intestinal epithelial cells, immunolocalization of the Mi-CK isozyme indicates that it is associated with long, thin mitochondria, which are excluded from the brush border at the apical end of each cell. In contrast, immunolocalization of the B-CK isozyme indicates that it is concentrated distinctly in the brush border terminal web domain. Although absent from the microvilli, B-CK also is distributed diffusely throughout the cytoplasm. Terminal web localization of B-CK was maintained in glycerol-permeabilized cells and in isolated brush borders, indicating that B-CK binds to the brush border structure. The abundance and localization of the mitochondrial and cytoplasmic creatine kinase isozymes suggest that they are part of a system that temporally and/or spatially buffers dynamic energy requirements of intestinal epithelial cells.     

Induction of p18INK4c and its predominant association with CDK4 and CDK6 during myogenic differentiation.

Terminal cell differentiation involves permanent withdrawal from the cell division cycle. The inhibitors of cyclin-dependent kinases (CDKs) are potential molecules functioning to couple cell cycle arrest and cell differentiation. In murine C2C12 myoblast cells, G1 CDK enzymes (CDK2, CDK4, and CDK6) associate with four CDK inhibitors: p18INK4c, p19INK4d, p21, and p27Kip1. During induced myogenesis, p21 and its associated CDK proteins underwent an initial increase followed by a decrease as cells became terminally differentiated. The level of p27 protein gradually increased, but the amount of total associated CDK proteins remained unchanged. p19 protein decreased gradually during differentiation, as did its associated CDK4 protein. In contrast, p18 protein increased 50-fold, from negligible levels in proliferating myoblasts to clearly detectable levels within 8-12 h of myogenic induction. This initial rise was followed by a precipitous increase between 12 and 24 h postinduction, with p18 protein finally accumulating to its highest level in terminally differentiated cells. Induction of p18 correlated with increased and sequential complex formation--first increasing association with CDK6 and then with CDK4 over the course of myogenic differentiation. All of the CDK6 and half of the CDK4 were complexed with p18 in terminally differentiated C2C12 cells as well as in adult mouse muscle tissue. Finally, kinase activity of CDK2 and CDK4 decreases as C2C12 cells differentiate, whereas the CDK6 kinase activity is low in both proliferating myoblasts and differentiated myotubes. Our results indicate that p18 may play a critical role in causing and/or maintaining permanent cell cycle arrest associated with mature muscle formation.     

Analysis of the myogenic lineage in chick embryos: I. Studies on the terminal cell division

An antibody prepared against the MM isozyme of creatine phosphokinase (M-CK) stained multinucleated myotubes and post-mitotic mononucleated myoblasts in mass cultures of myogenic cells taken from the breast muscles of 11-day chick embryos. No cycling cells bound the antibody. Single cells isolated either directly from the embryo or from mass cultures were seeded at clonal density and allowed to undergo one division. The resulting pairs of cells were stained with the antibody and were scored as (a) both members of the pair M-CK⁺; (b) both M-CK^?; or (c) mixed (one M-CK⁺ and one M-CK^?). No mixed pairs were observed. Conditioned medium did not induce all myogenic pairs to differentiate and growth medium did not keep myogenic pairs in the cell cycle. About 10% of clonal pairs established from 10 h cultures were M-CK⁺, while about 27% of pairs established from 30 h mass cultures were M-CK⁺. These results indicate that (1) the myogenic lineage ends in a symmetrical division whose products are two post-mitotic M-CK⁺ cells; (2) the expression of the muscle phenotype is not determined exclusively by the environment; (3) the terminal cells are the product of an intrinsic program or cell lineage in which only the last cells can synthesize muscle-specific proteins.     

Inflammatory cytokines inhibit myogenic differentiation through activation of nuclear factor-kappaB.

Muscle wasting is often associated with chronic inflammation. Because tumor necrosis factor alpha (TNF-alpha) has been implicated as a major mediator of cachexia, its effects on C2C12 myocytes were examined. TNF-alpha activated nuclear factor-kappaB (NF-kappaB) and interfered with the expression of muscle proteins in differentiating myoblasts. Introduction of a mutant form of inhibitory protein kappaBalpha (IkappaBalpha) restored myogenic differentiation in myoblasts treated with TNF-alpha or interleukin 1beta. Conversely, activation of NF-kappaB by overexpression of IkappaB kinase was sufficient to block myogenesis, illustrating the causal link between NF-kappaB activation and inhibition of myogenic differentiation. The inhibitory effects of TNF-alpha on myogenic differentiation were reversible, indicating that the effects of the cytokine were not due to nonspecific toxicity. Treatment of differentiated myotubes with TNF-alpha did not result in a striking loss of muscle-specific proteins, which shows that myogenesis was selectively affected in the myoblast stage by TNF-alpha. An important finding was that NF-kappaB was activated to the same extent in differentiating and differentiated cells, illustrating that once myocytes have differentiated they become refractory to the effects of NF-kappaB activation. These results demonstrate that inflammatory cytokines may contribute to muscle wasting through the inhibition of myogenic differentiation via a NF-kappaB-dependent pathway.     

Inhibition of myogenic differentiation by fibroblast growth factor or type beta transforming growth factor does not require persistent c-myc expression

Skeletal muscle differentiation is accompanied by accumulation of the mRNA encoding the muscle isoenzyme of creatine kinase (MCK) and can be suppressed by serum components, fibroblast growth factor (FGF), or type beta transforming growth factor (TGF beta). Using the nonfusing myogenic cell line, BC3H1, the potential involvement of c-myc in growth factor-dependent inhibition of myogenesis was examined. Withdrawal of undifferentiated myoblasts from the cell cycle in medium with 0.5% serum was associated with a precipitous decline in expression of c-myc mRNA followed by induction of MCK mRNA. In 0.5% serum containing TGF beta, c-myc mRNA declined to a level identical to that in differentiated cells; however, MCK mRNA was not expressed. Exposure of quiescent differentiated cells to FGF or TGF beta caused disappearance of muscle-specific gene products and was accompanied by only transient low level induction of c-myc mRNA. These data indicate that persistent c-myc expression is not required for growth factor-mediated inhibition of myogenic differentiation.