Muscle cell components dictate hepatocyte gene expression and the distribution of the Golgi apparatus in heterokaryons

The myonuclear domain in adult skeletal muscle fibres: past, present and future

Most cells in the body are mononuclear whereas skeletal muscle fibres are uniquely multinuclear. The nuclei of muscle fibres (myonuclei) are usually situated peripherally which complicates the equitable distribution of gene products. Myonuclear abundance can also change under conditions such as hypertrophy and atrophy. Specialised zones in muscle fibres have different functions and thus distinct synthetic demands from myonuclei. The complex structure and regulatory requirements of multinuclear muscle cells understandably led to the hypothesis that myonuclei govern defined 'domains' to maintain homeostasis and facilitate adaptation. The purpose of this review is to provide historical context for the myonuclear domain and evaluate its veracity with respect to mRNA and protein distribution resulting from myonuclear transcription. We synthesise insights from past and current in vitro and in vivo genetically modified models for studying the myonuclear domain under dynamic conditions. We also cover the most contemporary knowledge on mRNA and protein transport in muscle cells. Insights from emerging technologies such as single myonuclear RNA-sequencing further inform our discussion of the myonuclear domain. We broadly conclude: (1) the myonuclear domain can be flexible during muscle fibre growth and atrophy, (2) the mechanisms and role of myonuclear loss and motility deserve further consideration, (3) mRNA in muscle is actively transported via microtubules and locally restricted, but proteins may travel far from a myonucleus of origin and (4) myonuclear transcriptional specialisation extends beyond the classic neuromuscular and myotendinous populations. A deeper understanding of the myonuclear domain in muscle may promote effective therapies for ageing and disease.

The Ciona myogenic regulatory factor functions as a typical MRF but possesses a novel N-terminus that is essential for activity

Electroporation-based assays were used to test whether the myogenic regulatory factor (MRF) of Ciona intestinalis (CiMRF) interferes with endogenous developmental programs, and to evaluate the importance of its unusual N-terminus for muscle development. We found that CiMRF suppresses both notochord and endoderm development when it is expressed in these tissues by a mechanism that may involve activation of muscle-specific microRNAs. Because these results add to a large body of evidence demonstrating the exceptionally high degree of functional conservation among MRFs, we were surprised to discover that non-ascidian MRFs were not myogenic in Ciona unless they formed part of a chimeric protein containing the CiMRF N-terminus. Equally surprising, we found that despite their widely differing primary sequences, the N-termini of MRFs of other ascidian species could form chimeric MRFs that were also myogenic in Ciona. This domain did not rescue the activity of a Brachyury protein whose transcriptional activation domain had been deleted, and so does not appear to constitute such a domain. Our results indicate that ascidians have previously unrecognized and potentially novel requirements for MRF-directed myogenesis. Moreover, they provide the first example of a domain that is essential to the core function of an important family of gene regulatory proteins, one that, to date, has been found in only a single branch of the family.

Tumorigenic potential is restored during differentiation in fusion-reprogrammed cancer cells

Detailed understanding of the mechanistic steps underlying tumor initiation and malignant progression is critical for insights of potentially novel therapeutic modalities. Cellular reprogramming is an approach of particular interest because it can provide a means to reset the differentiation state of the cancer cells and to revert these cells to a state of non-malignancy. Here, we investigated the relationship between cellular differentiation and malignant progression by the fusion of four independent mouse cancer cell lines from different tissues, each with differing developmental potentials, to pluripotent mouse embryonic stem (ES) cells. Fusion was accompanied by loss of differentiated properties of the four parental cancer cell lines and concomitant emergence of pluripotency, demonstrating the feasibility to reprogram the malignant and differentiative properties of cancer cells. However, the original malignant and differentiative phenotypes re-emerge upon withdrawal of the fused cells from the embryonic environment in which they were maintained. cDNA array analysis of the malignant hepatoma progression implicated a role for Foxa1, and silencing Foxa1 prevented the re-emergence of malignant and differentiation-associated gene expression. Our findings support the hypothesis that tumor progression results from deregulation of stem cells, and our approach provides a strategy to analyze possible mechanisms in the cancer initiation.

Molecular roadblocks for cellular reprogramming

During development, diverse cellular identities are established and maintained in the embryo. Although remarkably robust in vivo, cellular identities can be manipulated using experimental techniques. Lineage reprogramming is an emerging field at the intersection of developmental and stem cell biology in which a somatic cell is stably reprogrammed into a distinct cell type by forced expression of lineage-determining factors. Lineage reprogramming enables the direct conversion of readily available cells from patients (such as skin fibroblasts) into disease-relevant cell types (such as neurons and cardiomyocytes) or into induced pluripotent stem cells. Although remarkable progress has been made in developing novel reprogramming methods, the efficiency and fidelity of reprogramming need to be improved in order increase the experimental and translational utility of reprogrammed cells. Studying the mechanisms that prevent successful reprogramming should allow for improvements in reprogramming methods, which could have significant implications for regenerative medicine and the study of human disease. Furthermore, lineage reprogramming has the potential to become a powerful system for dissecting the mechanisms that underlie cell fate establishment and terminal differentiation processes. In this review, we will discuss how transcription factors interface with the genome and induce changes in cellular identity in the context of development and reprogramming.

Maintaining differentiated cellular identity

Various studies have demonstrated that somatic differentiated cells can be reprogrammed into other differentiated states or into pluripotency, thus showing that the differentiated cellular state is not irreversible. These findings have generated intense interest in the process of reprogramming and in mechanisms that govern the pluripotent state. However, the realization that differentiated cells can be triggered to switch to considerably different lineages also emphasizes that we need to understand how the identity of mature cells is normally maintained. Here we review recent studies on how the differentiated state is controlled at the transcriptional level and discuss how new insights have begun to elucidate mechanisms underlying the stable maintenance of mature cell identities.

Reprogramming mediated by cell fusion technology

This review is focused on recent advances in fusion-based reprogramming of cells of different pluripotent statuses or lineage origins. Recent findings are discussed from standpoints of both the developmental potency of hybrid cells generated by fusion of pluripotent embryonic stem (ES) cells, embryonal carcinoma (EC) cells, and somatic cells and epigenetic mechanisms and other aspects involved in the reprogramming process. Complete reprogramming occurs at least 5-7 days after fusion and includes at least two steps. (i) initiation at the heterokaryon stage and choice of the direction of reprogramming using an "all-or-none principle" to establish the dominance of one parental genome and (ii) "fixation" of the newly acquired expression profile by epigenetic mechanisms. The first step is realized without cell division, whereas the second requires cell proliferation. Reprogramming in hybrid cells is rapid and complete. Thus, cell fusion is a powerful tool for reprogramming.

Nuclear reprogramming to a pluripotent state by three approaches

The stable states of differentiated cells are now known to be controlled by dynamic mechanisms that can easily be perturbed. An adult cell can therefore be reprogrammed, altering its pattern of gene expression, and hence its fate, to that typical of another cell type. This has been shown by three distinct experimental approaches to nuclear reprogramming: nuclear transfer, cell fusion and transcription-factor transduction. Using these approaches, nuclei from 'terminally differentiated' somatic cells can be induced to express genes that are typical of embryonic stem cells, which can differentiate to form all of the cell types in the body. This remarkable discovery of cellular plasticity has important medical applications.

Nuclear reprogramming to a pluripotent state by three approaches

The stable states of differentiated cells are now known to be controlled by dynamic mechanisms that can easily be perturbed. An adult cell can therefore be reprogrammed, altering its pattern of gene expression, and hence its fate, to that typical of another cell type. This has been shown by three distinct experimental approaches to nuclear reprogramming: nuclear transfer, cell fusion and transcription-factor transduction. Using these approaches, nuclei from 'terminally differentiated' somatic cells can be induced to express genes that are typical of embryonic stem cells, which can differentiate to form all of the cell types in the body. This remarkable discovery of cellular plasticity has important medical applications.

H3K27 trimethylation is an early epigenetic event of p16INK4a silencing for regaining tumorigenesis in fusion reprogrammed hepatoma cells

Stable epigenetic silencing of p16(INK4a) is a common event in hepatocellular carcinoma (HCC) cells, which is associated with abnormal cell proliferation and liberation from cell cycle arrest. Understanding the early epigenetic events in silencing p16(INK4a) expression may illuminate a prognostic strategy to block HCC development. Toward this end, we created a reprogram cell model by the fusion mouse HCC cells with mouse embryonic stem cells, in which the ES-Hepa hybrids forfeited HCC cell characteristics along with reactivation of the silenced p16(INK4a). HCC characteristics, in terms of gene expression pattern and tumorigenic potential, was restored upon induced differentiation of these reprogrammed ES-Hepa hybrids. The histone methylation pattern relative to p16(INK4a) silencing during differentiation of the ES-Hepa hybrids was analyzed. H3K27 trimethylation at the p16(INK4a) promoter region, occurring in the early onset of p16(INK4a) silencing, was followed by H3K9 dimethylation at later stages. During the induced differentiation of the ES-Hepa hybrids, H3K4 di- and trimethylations were maintained at high levels during the silencing of p16(INK4a), strongly suggesting that H3K4 methylation events did not cause the silencing of p16(INK4a). Our results suggested that the enrichment of H3K27 trimethylation, independent of H3K9 dimethylation, trimethylation, and DNA methylation, was an early event in the silencing of p16(INK4a) during the tumor development. This unique chromatin pattern may be a heritable marker of epigenetic regulation for p16(INK4a) silencing during the developmental process of hepatocellular carcinogenesis.

Acquisition and extinction of gene expression programs are separable events in heterokaryon reprogramming

Although differentiated cells normally retain cell-type-specific gene expression patterns throughout their lifetime, cell identity can sometimes be modified or reversed in vivo by transdifferentiation, or experimentally through cell fusion or by nuclear transfer. To examine the epigenetic changes that are required for the dominant conversion of lymphocytes to muscle, we generated heterokaryons between human B lymphocytes and mouse C2C12 myotubes. We show that within 2 days of heterokaryon formation lymphocyte nuclei adopt an architecture resembling that of muscle and then initiate the expression of muscle-specific genes in the same temporal order as developing muscle. The establishment of this muscle-specific program is coordinated with the shutdown of several lymphocyte-associated genes. Interestingly, erasing lymphocyte identity in reprogrammed cells requires histone deacetylase (HDAC) activity. Inhibition of HDAC activity during reprogramming selectively blocks the silencing of lymphocyte-specific genes but does not prevent the establishment of muscle-specific gene expression. Successful reprogramming is therefore shown to be a multi-step process in which the acquisition and extinction of lineage-specific gene programs are separable events.

Plasticity of cell fate: insights from heterokaryons

Experiments with somatic cell hybrids and stable heterokaryons have demonstrated that differentiated cells exhibit a remarkable capacity to change. Heterokaryons have been particularly useful in determining the extent to which the differentiated state of a cell is plastic. Cell fate can be altered by a change in the balance of positive and negative trans-acting regulators. Although a single regulator may be sufficient in certain environments to trigger a change in cell fate, that regulator may be ineffective in other cell contexts where it encounters a different composition of regulators.

High-efficiency retroviral infection of primary myoblasts

In the past, it has been hard to introduce genes into primary myoblasts without selection, as they have been very difficult to transfect or infect. We describe conditions under which mouse primary skeletal muscle myoblasts can be infected with retroviral vectors at high efficiency. Infection can be greatly increased by minimizing the time during which cells are exposed to virus, adding a minimal centrifugation step, and supplementing the infection cocktail to mimic more closely primary myoblast growth medium. Under these conditions, one round of exposure to virus results in an infection efficiency of up to 80%, whereas 4-5 rounds of infection over a two day period reproducibly yield an infection efficiency of > 99%. These methods greatly enhance the potential for studying genetically engineered primary myoblasts from any mouse strain, transgenic or knockout, and may have useful application to other primary cell types that are refractory to transfection or infection.

Regionalized expression of myosin isoforms in heterotypic myotubes formed from embryonic and fetal rat myoblasts in vitro

The development of mammalian limb muscles involves the appearance and fusion of at least two separate populations of muscle precursor cells. These two populations, termed embryonic and fetal myoblasts, are first detected within the limb bud at different stages of development. We have previously demonstrated that, in the rat, each myoblast population expresses a unique pattern of myosin heavy chains (MyHCs) during differentiation in vitro (Pin and Merrifield [1993] Dev. Genet. 14:356-368). Embryonic myoblasts accumulate embryonic and slow MyHCs, whereas fetal myoblasts accumulate embryonic, neonatal, and adult fast MyHCs but not slow MyHC. To determine if the two populations can fuse with each other and whether the pattern of MyHC expression is altered in the resulting heterokaryons, embryonic and fetal myoblasts were labelled with the lipophilic dye PKH26, [3H]-thymidine, or 5-bromodeoxyuridine (BRDU) and cocultured for 24-48 hr. Our results demonstrate that fusion occurs between embryonic and fetal myoblasts in vitro. Moreover, analysis of the resulting heterokaryons revealed regionalized accumulations of MyHC around individual nuclei. Interestingly, these accumulations were typical of the default pattern of expression that individual nuclei would have normally expressed in single culture. Nuclei contributed by embryonic myoblasts were surrounded by localized accumulations of slow MyHC, whereas nuclei from fetal myoblasts were surrounded by neonatal/fast MyHC. The occurrence of such nuclear domains indicates that the myoblast-specific expression of MyHC isoforms is dictated by cis-acting factors established prior to fusion.

Defective myogenesis in NFB-s mutant associated with a saturable suppression of MYF5 activity

Myogenic cell lines have proved to be useful tools for investigating the molecular mechanisms that control cellular differentiation. NFB-s is a mutant myogenic cell line which fails to differentiate in vitro, and can repress differentiation in normal myogenic cells when fused to form heterokaryons. The NFB-s cell line was used here to study the molecular mechanisms underlying such myogenic repression. Using muscle-specific reporter genes, we show that NFB-s cells fail to activate fully the muscle differentiation program at a transcriptional level, although muscle-specific transcription can be enhanced by regulators of differentiation such as pertussis toxin. Paradoxically we find that the myogenic regulator myf5 is expressed at constitutively high levels in NFB-s cells, and retains DNA binding activity. Expression plasmids encoding NFB-derived myf5 cDNA can rescue the myogenic phenotype in NFB-s cells, demonstrating that a threshold level of positive regulators must be reached before the myogenic program is activated. Thus, the dominant negative phenotype does not appear to result from defective myf5, but is due to a dosage-dependent saturable mechanism that interferes with myf5 function. These studies demonstrate that the stoichiometric ratio of positive and negative regulators is critical for determining the myogenic differentiation state.

Coordinate suppression of myeloma-specific genes and expression of fibroblast-specific genes in myeloma X fibroblast somatic cell hybrids

In most instances, fusion of differentiated cell types with fibroblasts has resulted in the extinction of the differentiation-specific traits of the non-fibroblast parental cell. To explore the genetic basis of this phenomenon, we have studied a series of somatic cell hybrids between mouse myeloma and fibroblasts. All the hybrids were adherent having a fibroblast-like phenotype. Molecular analysis revealed that plasma cell specific genes like the productively rearranged Ig genes, the J chain gene and genes for the cell surface markers CD20 and PC1, were extinguished in the hybrids. In contrast, fibroblast specific genes like fibronectin, alpha 2(I) and III collagens, as well as the receptor for fibroblast growth factor (flg), were expressed. Extinction was not due to chromosomal loss or lack of the relevant genes. To learn about the mechanism(s) of this phenomenon we have looked for the presence of positive and negative transcription factors in our hybrids. Expression of the PU.1 transcription factor, a member of the Ets transcription factor family normally expressed in B cells and macrophages, was lost in the cell hybrids. Interestingly, we found that the B-cell-specific Oct-2 transcription factor was still expressed at somewhat variable levels in several of the hybrid cell lines. In contrast, expression of the recently identified octamer coactivator BOB.1/OBF.1 was extinguished in all cell hybrids. This supports a critical role of this transcriptional coactivator for B-cell-specific gene expression. In addition, the Id and HLH462 genes coding for proteins known to repress bHLH transcription factors by formation of heterodimers, were found to be expressed at increased levels in fibroblasts and in the hybrids, indicating that their increased levels might also contribute to the suppression of myeloma-specific genes. Our results show that in myeloma x fibroblast hybrids, the phenotype of the fibroblast is dominant. It is suggested that fibroblasts contain regulatory "master" genes that are responsible for activation of the fibroblast differentiation pathway and suppress differentiation programs of other cell types.

Developmental regulation of membrane traffic organization during synaptogenesis in mouse diaphragm muscle

In innervated adult skeletal muscles, the Golgi apparatus (GA) displays a set of remarkable features in comparison with embryonic myotubes. We have previously shown by immunocytochemical techniques, that in adult innervated fibers, the GA is no longer associated with all the nuclei, but appears to be concentrated mostly in the subneural domain under the nerve endings in chick (Jasmin, B. J., J. Cartaud, M. Bornens, and J.-P. Changeux. 1989. Proc. Natl. Acad. Sci. USA. 86:7218-7222) and rat (Jasmin, B. J., C. Antony, J.-P. Changeux, and J. Cartaud. 1995. Eur. J. Neurosci. 7:470-479). In addition to such compartmentalization, biochemical modifications take place that suggest a functional specialization of the subsynaptic GA. Here, we focused on the developmental regulation of the membrane traffic organization during the early steps of synaptogenesis in mouse diaphragm muscle. We investigated by immunofluorescence microscopy on cryosections, the distribution of selected subcompartments of the exocytic pathway, and also of a representative endocytic subcompartment with respect to the junctional or extrajunctional domains of developing myofibers. We show that throughout development the RER, the intermediate compartment, and the prelysosomal compartment (mannose 6-phosphate receptor-rich compartment) are homogeneously distributed along the fibers, irrespective of the subneural or extrajunctional domains. In contrast, at embryonic day E17, thus 2-3 d after the onset of innervation, most GA markers become restricted to the subneural domain. Interestingly, some Golgi markers (e.g., alpha-mannosidase II, TGN 38, present in the embryonic myotubes) are no longer detected in the innervated fiber even in the subsynaptic GA. These data show that in innervated muscle fibers, the distal part of the biosynthetic pathway, i.e., the GA, is remodeled selectively shortly after the onset of innervation. As a consequence, in the innervated fiber, the GA exists both as an evenly distributed organelle with basic functions, and as a highly differentiated subsynaptic organelle ensuring maturation and targeting of synaptic proteins. Finally, in the adult, denervation of a hemidiaphragm causes a burst of reexpression of all Golgi markers in extrasynaptic domains of the fibers, hence showing that the particular organization of the secretory pathway is placed under nerve control.

Nerve-dependent plasticity of the Golgi complex in skeletal muscle fibres: compartmentalization within the subneural sarcoplasm

Several recent reports have highlighted the plasticity of the Golgi apparatus during myogenesis, yet the organization of this specialized organelle in innervated skeletal muscle fibres remains poorly understood. Using four bona fide anti-Golgi antibodies, directed against a 210 kDa protein, a 160 kDa sialoglycoprotein, the small GTP-binding protein rab6p, and TGN38, the localization of which covers the various compartments of the Golgi complex, we show by immunofluorescence microscopy that the Golgi complex undergoes considerable reorganization in the course of myogenic differentiation and motor endplate formation in the rat. Unlike the typical perinuclear distribution of the Golgi stacks associated with every nucleus in myotubes, a striking subneural compartmentalization is observed in adult innervated myofibres. In short-term denervated adult muscle fibres, we noticed the presence of the perinuclear Golgi apparatus in extrajunctional regions, a pattern reminiscent of that of developing myotubes. At variance with anti-Golgi antibodies, antibodies to the rough endoplasmic reticulum label structures dispersed throughout the entire sarcoplasm, hence suggesting that it is not the entire membrane/secretory protein synthesis machinery which is compartmentalized, but only the Golgi apparatus. Also, an unexpected lack of immunoreactivity with the TGN38 and alpha-mannosidase II antibodies points to biochemical differentiation of the subneural Golgi apparatus at the adult motor endplate. These new data extend our previous observations on the compartmentalization of the Golgi apparatus in the postsynaptic sarcoplasm of chick muscle fibres, and further illustrate the plasticity of the Golgi apparatus in muscle cells. The specialization of the Golgi apparatus within the subneural compartment provides this particular region with a compartmentalized secretory pathway, and these observations highlight the notion that the level of differentiation of this domain is not only maintained via transcriptional regulation but also by post-translational control mechanisms.

Heterokaryons of cardiac myocytes and fibroblasts reveal the lack of dominance of the cardiac muscle phenotype

The molecular characterization of a cardiac determination gene has been an elusive goal for the past several years. Prior to cloning of the skeletal muscle determination factor MyoD, the presence of a dominantly acting skeletal muscle determination factor had been inferred from the observation that the skeletal muscle phenotype was dominant in skeletal muscle-fibroblast heterokaryons (H. M. Blau, G. K. Pavlath, E. C. Hardeman, C.-P. Chiu, L. Siberstein, S. G. Webster, S. C. Miller, and D. Webster, Science 230:758-766, 1985). In these experiments, we have examined cardiac-fibroblast heterokaryons to investigate the existence of a dominantly acting cardiac determination factor. We have employed a novel experimental approach using primary embryonic fibroblasts from transgenic mice as a means of assaying for the activation of a cardiac promoter-luciferase reporter transgene within fibroblast nuclei. This approach provides a potential means of genetic selection for a dominantly acting positive factor and can be generalized to other systems. We have examined the expression of three markers of the cardiac lineage: a myofibrillar protein promoter (MLC2), a secreted protein (ANF), and a transcription factor (MEF2). MEF2 is specific to both cardiac and skeletal muscle cells. Our results indicate that in a majority of heterokaryons with an equal ratio of cardiac to fibroblast nuclei, none of these cardiac markers are expressed, indicating that the cardiac phenotype is not dominant over the embryonic fibroblast phenotype. The distinction from previous results with skeletal muscle is emphasized by our results with MEF2, which is dominantly expressed in skeletal muscle-fibroblast but not cardiac-fibroblast heterokaryons, supporting its divergent regulation in the two cell types.

Heterokaryons of cardiac myocytes and fibroblasts reveal the lack of dominance of the cardiac muscle phenotype

The molecular characterization of a cardiac determination gene has been an elusive goal for the past several years. Prior to cloning of the skeletal muscle determination factor MyoD, the presence of a dominantly acting skeletal muscle determination factor had been inferred from the observation that the skeletal muscle phenotype was dominant in skeletal muscle-fibroblast heterokaryons (H. M. Blau, G. K. Pavlath, E. C. Hardeman, C.-P. Chiu, L. Siberstein, S. G. Webster, S. C. Miller, and D. Webster, Science 230:758-766, 1985). In these experiments, we have examined cardiac-fibroblast heterokaryons to investigate the existence of a dominantly acting cardiac determination factor. We have employed a novel experimental approach using primary embryonic fibroblasts from transgenic mice as a means of assaying for the activation of a cardiac promoter-luciferase reporter transgene within fibroblast nuclei. This approach provides a potential means of genetic selection for a dominantly acting positive factor and can be generalized to other systems. We have examined the expression of three markers of the cardiac lineage: a myofibrillar protein promoter (MLC2), a secreted protein (ANF), and a transcription factor (MEF2). MEF2 is specific to both cardiac and skeletal muscle cells. Our results indicate that in a majority of heterokaryons with an equal ratio of cardiac to fibroblast nuclei, none of these cardiac markers are expressed, indicating that the cardiac phenotype is not dominant over the embryonic fibroblast phenotype. The distinction from previous results with skeletal muscle is emphasized by our results with MEF2, which is dominantly expressed in skeletal muscle-fibroblast but not cardiac-fibroblast heterokaryons, supporting its divergent regulation in the two cell types.

Reversibility of the differentiated state in somatic cells

Analysis of de novo gene activation in multinucleated heterokaryons has shown that the differentiated state, although stable, is not irreversible, and can be reprogrammed in the presence of appropriate combinations of trans-acting regulatory molecules. These properties have been exploited to design strategies for identifying novel regulators of cellular differentiation.

Restricted distribution of mRNA produced from a single nucleus in hybrid myotubes

Although the proteins encoded by a single nucleus in multinucleated myotubes have a wide range of distributions within the myofiber, little is known about the distributions of their mRNAs. We have used hybrid myotubes in which one or a few nuclei are derived from myoblasts that express nonmuscle proteins to investigate this question. We find that three different mRNAs, encoding proteins that are, respectively, nuclear, cytoplasmic, and targeted to the ER, have similar distributions within myotubes. Each is confined to an area within approximately 100 microns of the nucleus that expresses it.

The germ layer origin of mouse vaginal epithelium restricts its responsiveness to mesenchymal inductors: uterine induction

The epithelium of the mammalian vagina arises from two distinct germ layers, endoderm from the urogenital sinus and mesoderm from the lower fused Müllerian ducts. While previously it has been reported that neonatal vaginal epithelium can be induced to differentiate as uterus, which normally develops from the middle portion of the Müllerian ducts, it has not been determined whether this ability is shared by both mesoderm- and endoderm-derived vaginal epithelia. To test if germ layer origin influences the ability of vaginal epithelium to undergo uterine differentiation, we have isolated sinus-derived and Müllerian-derived vaginal epithelia from newborn mice, combined them with uterine mesenchyme, and grown them for 4 weeks in female mice. Mesoderm-derived Müllerian vaginal epithelium in combination with uterine mesenchyme formed the simple columnar epithelium typical of uterus. Similar results were obtained with neonatal cervical epithelium, another mesodermal Müllerian duct derivative. On the other hand, sinus vaginal epithelium combined with uterine mesenchyme formed small cysts lined by a stratified squamous vaginal-like epithelium. This epithelium never showed evidence of cycling between the cornified and mucified states as is typically seen in vaginal epithelium combined with vaginal stroma. These results indicate that the ability of epithelium to form uterus is limited to mesoderm-derived epithelia and suggest that endoderm-derived sinus vaginal epithelium cannot undergo the typical differentiative modifications in response to the hormonal fluctuations of the estrous cycle when associated with uterine stroma.

Systemic delivery of human growth hormone by injection of genetically engineered myoblasts

A recombinant gene encoding human growth hormone (hGH) was stably introduced into cultured myoblasts with a retroviral vector. After injection of genetically engineered myoblasts into mouse muscle, hGH could be detected in serum for 3 months. The fate of injected myoblasts was assessed by coinfecting the cells with two retroviral vectors, one encoding hGH and the other encoding beta-galactosidase from Escherichia coli. These results provide evidence that myoblasts, which can fuse into preexisting multinucleated myofibers that are vascularized and innervated, may be advantageous as vehicles for systemic delivery of recombinant proteins.

The response of female urogenital tract epithelia to mesenchymal inductors is restricted by the germ layer origin of the epithelium: prostatic inductions

The epithelium of the mammalian vagina arises from two distinct germ layers, endoderm from the urogenital sinus and mesoderm from the Müllerian ducts. While neonatal vaginal epithelium can be induced to form prostate which is normally an endodermal derivative, it has not been determined whether this ability to form prostate is shared by both mesoderm- and endoderm-derived vaginal epithelia. To test the competence of vaginal epithelia we have isolated sinus-derived and Müllerian-derived vaginal epithelia from newborn mice, combined them with rat urogenital sinus mesenchyme, and grown the tissue recombinants for 4 weeks in male athymic nude mice. Endoderm-derived sinus vaginal epithelium was induced to form prostatic tissue which expressed prostate-specific secretory proteins in 21 of 23 tissue recombinants. Müllerian-derived vaginal epithelium formed small ducts and cysts lined by a simple epithelium. These latter tissue recombinants lacked any evidence of prostatic secretory proteins. Similarly, endoderm-derived urethral epithelium was induced to form prostate (17 of 17 cases), while mesoderm-derived uterine epithelium was not (0 of 13 cases). Therefore, the ability to form prostatic epithelium was limited to endodermal derivatives of the urogenital tract.

Transactivation of pancreas-specific gene sequences in somatic cell hybrids

Enhancer/promoter elements from two pancreas-specific genes, those encoding amylase and elastase, were ligated to the bacterial GPT gene. The resulting construct can be used to select for expression of gene products which activate these pancreas-specific promoters in hybrid cells. The selectable GPT construct was stably transferred into several cell lines either directly or by cotransfection with pSV2Neo. GPT was expressed when transferred to pancreatic cell lines but not when transferred to GPT-fibroblast (L) cells or hepatoma cells. When the transformed L cells and hepatoma cells were fused with pancreatic cell lines, GPT was activated in the hybrid cells. Endogenous pancreas-specific genes from the L-cell and hepatoma parents were also activated in the hybrids. In addition, a pancreas-specific nuclear protein, PTF1, was produced in pancreatic and hybrid cells, correlating with GPT expression. The transformed L cells and hepatoma cells thus contained a nonexpressed construct which could be activated in trans by factors present in pancreatic cells. The hepatoma hybrid also continued to produce albumin, demonstrating the coexpression of liver and pancreas-specific genes in the hybrid-cell population. Cell lines carrying the amylase/elastase/GPT construct may be useful as a selection system for cloning of pancreatic transcription activators.

Transactivation of pancreas-specific gene sequences in somatic cell hybrids

Enhancer/promoter elements from two pancreas-specific genes, those encoding amylase and elastase, were ligated to the bacterial GPT gene. The resulting construct can be used to select for expression of gene products which activate these pancreas-specific promoters in hybrid cells. The selectable GPT construct was stably transferred into several cell lines either directly or by cotransfection with pSV2Neo. GPT was expressed when transferred to pancreatic cell lines but not when transferred to GPT-fibroblast (L) cells or hepatoma cells. When the transformed L cells and hepatoma cells were fused with pancreatic cell lines, GPT was activated in the hybrid cells. Endogenous pancreas-specific genes from the L-cell and hepatoma parents were also activated in the hybrids. In addition, a pancreas-specific nuclear protein, PTF1, was produced in pancreatic and hybrid cells, correlating with GPT expression. The transformed L cells and hepatoma cells thus contained a nonexpressed construct which could be activated in trans by factors present in pancreatic cells. The hepatoma hybrid also continued to produce albumin, demonstrating the coexpression of liver and pancreas-specific genes in the hybrid-cell population. Cell lines carrying the amylase/elastase/GPT construct may be useful as a selection system for cloning of pancreatic transcription activators.

Modulation of cellular glycosidase activity by hyperthermia

We examined the effect of 45 degrees C hyperthermia on the following glycosidases in CHO cells: beta-galactosidase, beta-hexosaminidase, beta-glucuronidase and alpha-mannosidase. Among these, lysosomal alpha-mannosidase exhibited the most dramatic response to hyperthermia with an increase in activity immediately after 45 degrees C hyperthermia. The increase was linearly dose-dependent with a doubling of activity for every 20 min at 45 degrees C. In contrast to alpha-mannosidase, beta-glucuronidase, beta-galactosidase, and beta-hexosaminidase showed only minor alterations in activity, or none, after hyperthermia of 10 to 60 min at 45 degrees C. Induction of thermotolerance enhanced the heat resistance of beta-galactosidase, but caused increased heat sensitivity for alpha-mannosidase. Intracellular beta-galactosidase, measured by histochemical staining, showed a dramatic redistribution in response to mild hyperthermia (10 min, 45 degrees C); the same effect was not observed for beta-glucuronidase. The data argue against non-specific activation of lysosomes by hyperthermia, and suggest that cells contain lysosomal subpopulations that are characterized by different heat sensitivities and variable glycosidase contents.

Reprogramming of myosin light chain 1/3 expression in muscle heterokaryons

Fast myosin light chain (MLC) 1/3 is one of the few genes which regulates transcript production at both transcriptional and post-transcriptional levels, utilizing two functionally distinct promoters coupled with alternatively spliced exons. The transcriptional process controlling expression from this single gene locus is developmentally regulated, such that MLC 1 precedes MLC 3 during myogenesis. Results from our RNA analyses demonstrate that in differentiated rat L6E9 muscle, MLC 3 is the sole isoform expressed from the MLC 1/3 locus. However, we also show that by generating rat L6E9:mouse C2 muscle heterokaryons, MLC 1 expression from the L6E9 MLC locus can be induced. In addition to novel rat MLC 1 expression in the C2:L6E9 heterokaryons, we show that the synthesis profile of rat MLC 3 mRNA is also altered relative to L6E9 muscle cultured alone. Additional experiments demonstrate that the reprogramming of rat MLC 1 and 3 expression in the muscle heterokaryons requires that C2 and L6E9 nuclei be contained within a common cytoplasm. These results demonstrate that expression from the MLC 1/3 gene is "plastic," and is not under the control of a strict developmental program but, rather, can be modified by the environmental milieu.

Extinction of retinol-binding protein gene expression in somatic cell-hybrids: identification of the target sequences

The Retinol-Binding Protein (RBP) is expressed primarily in the liver. The regulatory elements involved in its tissue-specific expression have been identified and mapped to the 5' flanking region of the RBP gene. In this paper heterokaryons and somatic cell-hybrids have been produced and analysed in order to demonstrate that the RBP gene is subject to extinction and to identify the target sequences of this phenomenon. We show here that the gene is extinguished in fusions of hepatoma with a variety of cells of different species and embryonic lineages. The repression is not due to loss of the gene and occurs also when chromosome 10, where the gene is located, is inherited from the expressing parental cell-type. Hybrid clones were transfected with constructs carrying DNA segments of different lengths from the 5' flanking region of the RBP gene fused to a reporter gene. We demonstrate that extinction takes place also on an exogenous RBP-CAT gene, mimicking the phenomenon observed with the endogenous gene in its chromosomal location. Moreover, we identify and map the target sequences of the putative extinguishing function. Our data thus show that extinction of RBP is mediated through the DNA segment that is involved in its tissue-specific expression.

Negative control of the helix-loop-helix family of myogenic regulators in the NFB mutant

We have characterized a nondifferentiating mouse muscle cell line, NFB, that represses the activity of the helix-loop-helix (HLH) family of myogenic regulators, yet expresses sarcomeric actins. The NFB MyoD gene is silent, but can be activated upon transfection of a long terminal region-controlled chicken MyoD cDNA, resulting in myogenesis. When NFB cells are fused with H9c2 rat muscle cells in heterokaryons, the level of rat MyoD transcripts declines. Thus, the stoichiometry of MyoD and the putative repressor controls myogenesis. Although NFB cells express myogenin and Myf-5 transcripts, the activity of these regulators is also repressed:myogenesis is not induced in 10T1/2 fibroblasts and is repressed in L6 muscle cells upon fusion with NFB cells. We conclude that the myogenic HLH regulators are not required for sarcomeric actin gene activation and that myogenesis is subject to dominant-negative control.

Effect of cell history on response to helix-loop-helix family of myogenic regulators

In multinucleated heterokaryons formed from the fusion of differentiated muscle cells to either hepatocytes or fibroblasts, muscle-specific gene expression is activated, liver-specific gene expression is repressed, and there are changes in the location of the Golgi apparatus. An understanding of the regulatory mechanisms that underlie this plasticity is of particular interest given the stability of the differentiated state in vivo. We have now investigated whether MyoD or myogenin, regulators of muscle-specific gene expression that have a helix-loop-helix motif, can induce the phenotypic conversion observed in heterokaryons. When these regulators were stably or transiently introduced into fibroblasts or hepatocytes by microinjection, transfection or retroviral infection with complementary DNA in expression vectors, fibroblasts expressed muscle-specific genes, whereas hepatocytes did not. However, fusion of hepatocytes stably expressing MyoD to fibroblasts resulted in activation in the heterokaryon of muscle-specific genes of both cell types. These results imply that other regulators, present in fibroblasts but not in hepatocytes, are necessary for the activation of muscle-specific genes, and indicate that the differentiated state of a cell is dictated by its history and a dynamic interaction among the proteins that it contains.

Localization of an acetylcholine receptor intron to the nuclear membrane

The first intron of the RNA for the acetylcholine receptor (AChR) alpha subunit shows a ringlike distribution around nuclei in multinucleated myotubes by in situ hybridization. This pattern is not observed for an actin intron or U1 RNA. Quantitation of the intron sequences reveals large variations in the amount of both the AChR and actin introns between nuclei within the same myotube, although all nuclei express equivalent amounts of U1 RNA. This differential RNA expression indicates that nuclei can individually control expression of messenger RNAs. The restricted distribution of the AChR intron RNA suggests a previously unknown step in RNA processing.

Localization of the sea urchin Spec3 protein to cilia and Golgi complexes of embryonic ectoderm cells

Expression of the Spec3 gene of Strongylocentrotus purpuratus is associated with ectodermal ciliogenesis. An antiserum was raised against the amino terminus of the deduced Spec3 amino acid sequence and used for immunofluorescent staining. Cilia and an apical structure at the base of the stained cilium of each ectodermal cell stained intensely in gastrula and later stage embryos. Microtubule-depolymerizing agents dispersed the concentrated spot of apical staining, suggesting a localization of Spec3 antigen to the Golgi complex. Immunogold electron microscopy confirmed the localization of Spec3 antigen on cilia and in the Golgi complex. Spec3 antigen showed a diffuse punctate staining pattern in the ectodermal cytoplasm of hatching blastula when Spec3 transcripts are most prevalent, suggesting that after synthesis, Spec3 is sequestered in the Golgi complex before appearing on cilia. Whereas the predicted Mr of the Spec3 protein is 21,600, immunoblotting with S. purpuratus proteins indicated that a Spec3 antigen was concentrated in cilia and migrated as an SDS-resistant aggregate of Mr approximately 350,000. Spec3 is also concentrated in cilia of Lytechinus pictus but the protein migrated with an Mr approximately 23,000 in this species. The S. purpuratus Spec3 antigen remains associated with the ciliary axoneme after extraction of membrane proteins.

Intracellular and surface distribution of a membrane protein (CD8) derived from a single nucleus in multinucleated myotubes

We have investigated the contribution of an individual nucleus to intracellular and surface membranes in multinucleated muscle fibers. Using a retroviral vector, we introduced the gene encoding the human T-lymphocyte antigen CD8 into C2 mouse muscle cells to form a stable line expressing the human protein on its surface. The intracellular and surface distributions of the protein were then investigated by immunocytochemistry in hybrid myotubes containing a single nucleus expressing CD8. We show that the intracellular distribution of CD8 is limited to a local area surrounding the nucleus encoding it and several neighboring nuclei. On the cell surface, however, the protein is distributed over the entire myotube. Widespread distribution of a surface membrane protein in multinucleated myotubes can thus result from localized synthesis and processing.

A myosin isoform repressed in hypertrophied ALD muscle of the chicken reappears during regeneration following cold injury

A library of monoclonal antibodies specific for myosin heavy chain (HC) was used to study myosin expression in regenerating fibers. The response to cold injury of slow skeletal ALD muscle previously induced to eliminate SM1 myosin by weight overload was compared to that of its contralateral control. Native gel electrophoresis combined with immunoblotting demonstrated that slow SM1 myosin HC eliminated from hypertrophic muscle reappeared both at the site of active regeneration and unexpectedly, also distal to the site of injury. The regeneration response of hypertrophied muscles was similar to that of the controls. In addition to SM1 myosin HC, ventricular-like and embryonic/fast isoforms were also expressed in both muscles during the early stages of regeneration and disappeared as the muscle fibers matured. These observations demonstrate that regenerating slow muscle fibers reexpress myosins' characteristic of developing muscle irrespective of the myosin phenotype prior to injury. The reappearance of repressed myosin HC in the hypertrophied ALD muscle is consistent with the presence of newly differentiated myonuclei.

Golgi apparatus in chick skeletal muscle: changes in its distribution during end plate development and after denervation

In the course of studies about the cellular and molecular mechanisms of motor end plate formation, the distribution of the Golgi apparatus (GA) has been investigated by immunofluorescence methods in chick skeletal muscle in primary culture and in innervated muscles of 15-day-old chicks. By using a monoclonal antibody directed against the GA, we confirmed the known distribution of the GA in myogenic cells: a juxtanuclear polarized organization in myoblasts and a perinuclear nonpolarized distribution in myotubes. In contrast, the innervated anterior latissimus dorsi muscle of "young adult" chicks displayed a focal distribution of GA that appeared restricted to areas located underneath the motor end plates identified by alpha-bungarotoxin fluorescent labeling of the acetylcholine receptor. Five days after denervation of anterior latissimus dorsi muscle, a striking reorganization and expansion of the GA was observed. The GA now showed a perinuclear distribution in close association with every nucleus of the muscle fibers as observed in myotubes. The focal distribution of the GA in innervated muscle fibers and its remodeling upon denervation are interpreted in terms of a model of local synthesis, processing, and routing of acetylcholine receptor to the end plate and of regulation of these processes by functional motor innervation.

How fixed is the differentiated state? Lessons from heterokaryons

The differentiated state is highly stable in vivo. Yet, in response to nuclear transplantation, tissue regeneration or cell fusion, the nuclei of differentiated cells exhibit a remarkable capacity to change. I review here the utility of heterokaryons, multinucleated cell hybrids, in elucidating the mechanisms that establish and maintain the differentiated state and yet allow such plasticity.

Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD

MyoD is a master regulatory gene for myogenesis. Under the control of a retroviral long terminal repeat, MyoD was expressed in a variety of differentiated cell types by using either a DNA transfection vector or a retrovirus. Expression of muscle-specific proteins was observed in chicken, human, and rat primary fibroblasts and in differentiated melanoma, neuroblastoma, liver, and adipocyte lines. The ability of MyoD to activate muscle genes in a variety of differentiated cell lines suggests that no additional tissue-specific factors other than MyoD are needed to activate the downstream program for terminal muscle differentiation or that, if such factors exist, they are themselves activated by MyoD expression.

In vivo aging of human fibroblasts does not alter nuclear plasticity in heterokaryons

In vivo aging of human fibroblasts altered proliferative properties but not the potential for novel gene expression in response to muscle trans-acting factors. Heterokaryons produced by fusing fibroblasts with muscle cells permitted a dissociation of the effects of aging on cell division and other cell functions. Skin fibroblasts derived from fetal and adult stages of development were distinct cell types based on their doubling time, protein content, cell size, and specific binding of insulin and insulin-like growth factor I. Despite these differences in growth parameters, the two cell types were indistinguishable in heterokaryons. Muscle gene activation occurred in the absence of changes in chromatin structure requiring DNA replication. In addition, the time course, maximal efficiency, and effect of gene dosage on the expression of muscle gene products were similar for heterokaryons containing fetal and adult fibroblasts but distinct for heterokaryons containing keratinocytes. The difference between fibroblasts and keratinocytes in the time course of muscle gene expression is likely to reflect mechanisms of gene activation at the transcriptional level, since the kinetics of muscle protein accumulation paralleled that of muscle transcripts. These results indicate that nuclear plasticity is not altered in fibroblasts by in vivo aging.

Localization of muscle gene products in nuclear domains

The localization of gene products is central to the development of cell polarity and pattern specification during embryogenesis. To monitor the distribution of gene products encoded by different nuclei in the same cell in tissue culture, we fused cells of different species to form multinucleated non-dividing heterokaryons. In previous fusion studies, cell-surface antigens and organelles contributed by disparate cell types intermixed within minutes. Using heterokaryons produced with differentiated muscle cells, we demonstrate here that a muscle membrane component, the Golgi apparatus mediating its transport, and a sarcomeric myosin heavy chain are localized in the vicinity of the nuclei responsible for their synthesis. These results provide direct evidence that products (organelle, membrane and structural proteins) derived from individual nuclei can remain localized in myotubes, a finding with implications both for neuromuscular synapse formation and for the carrier state of Duchenne muscular dystrophy.