Re-organization of nucleolar architecture in myogenic differentiation

The nucleolus, which was initially observed in the 1800s, is a phase separated subnuclear compartment primarily specialized for ribosome biogenesis (Pederson, 1998). Recognition that the nucleolus exhibits dynamic changes in its architecture depending on a variety of cellular conditions has been well documented (Boulon et al., 2010). Structurally, the clustered arrays of rRNA gene repeats that are sequestered and transcribed in the FC subdomain of the tri-partite nucleolar structure arise from five distinct acrocentric chromosomes in mouse and human. Remarkably, dynamic changes in the size and fusion of nucleoli within a given interphase nucleus depend on the reorganization of multiple chromosomal regions encoding the rDNA repeats (Lafontaine et al., 2021).

Clustering of the chromosomal domains encoding the rRNA genes nucleates the nucleolar organizer region (NOR), which seeds the formation of the nucleolus (Hernandez-Verdun, 2011). Actively transcribed rRNAs scaffold a variety of RNA-binding proteins that often contain intrinsically disordered regions (IDRs) responsible for weak multivalent protein–protein interactions (van der Lee et al., 2014). These assemblies lead to the formation of multiphase condensates characteristic of typical nucleolar structure. The core-shell nucleolar architecture thus consists of three layers or sub-compartments: the fibrillar center (FC); the dense fibrillar component (DFC); and the granular component (GC) (Feric et al., 2016). The FC forms the core containing rDNA repeats, transcription factors and RNA polymerase I (Pol I) for rRNA gene transcription. Nascent rRNAs are localized at the boundary between the FC and the DFC regions, and nascent rRNAs are sorted into the DFC and processed and modified by, for example, IDR-containing RNA-binding proteins, such as fibrillarin (Yao et al., 2019). In the outermost layer of the nucleolus, the GC, rRNAs and ribosomal proteins are assembled into rRNA–protein complexes (rRNPs) mediated by proteins such as pescadillo and nucleolin (Lerch-Gaggl et al., 2002; Tajrishi et al., 2011).

In proliferating cells, nucleolar formation is dynamically regulated (Hernandez-Verdun, 2006). Dissolution of the nucleolus occurs during mitosis, and the nucleolus is reassembled upon mitotic exit (Hernandez-Verdun, 2011). At the onset of mitosis, Pol I activity ceases and the loss of nascent rRNA results in dispersion of the FC and, subsequently, the DFC and the GC regions in a consecutive manner (Feric et al., 2016; Yao et al., 2019). The resumption of Pol I activity after exiting mitosis triggers the re-formation of the FC region, followed by the establishment of the FC, DFC and GC multilayered structure, which is characteristic of the mature interphase nucleolar architecture (Dundr et al., 2000; Hernandez-Verdun, 2011; Leung et al., 2004). Therefore, alterations in the structure of the nucleolus are intrinsically linked with changes in Pol I mediated transcription of rRNA genes and ultimately to a change in the protein translational capacity of the cell (Goodfellow and Zomerdijk, 2013). Thus, modulation of nucleolar morphology under a variety of cellular states may be functionally linked to adaptations in gene expression and cellular function.

In the process of muscle cell differentiation (myogenesis), the formation of multinucleated muscle cells (myotubes) from committed progenitor cells (myoblasts; MBs) requires a vast orchestration of gene expression. The myogenic gene program has been a paradigm for studies on cellular differentiation since the late 1980s when the MyoD gene was first identified and characterized as a master regulatory transcription factor of the myogenic program (Davis et al., 1987; Lassar et al., 1989; Weintraub et al., 1989). Since then, the identification of other members of this basic helix-loop-helix family of sequence specific DNA binding transcription factors (sTFs), collectively termed the myogenic regulatory factors [MRFs; MyoD (also known as MyoD1), Myf5, myogenin (MyoG) and MRF4 (also known as Myf6)] have been shown to play critical roles during myogenesis (Hasty et al., 1993; Kassar-Duchossoy et al., 2004; Nabeshima et al., 1993; Rudnicki et al., 1993). Further work documented the co-operative interaction of the MRFs with members of the myocyte enhancer factor 2 (MEF2) family, forming an anciently conserved bipartite determinant of muscle-specific gene transcription (Molkentin et al., 1995; Ornatsky et al., 1997). Thus, the transcriptional regulatory program underlying the production of Pol II-mediated mRNA transcriptome changes orchestrating myogenesis has been well characterized. Conversely, the large changes in protein synthesis necessary to translate the altered transcriptome at the inception of differentiation into changes in the proteome leading to differentiation are less well understood (Fujita et al., 2021, 2017; Zismanov et al., 2016). For example, the synthesis of large structural proteins for the bulk synthesis of sarcomeres requires substantial changes in translational control (Rommel et al., 2001).

In addition to proliferation and differentiation of myogenic cells, mature skeletal muscle fibers also possess substantial plasticity in response to a wide range of environmental and physiological demands (Egan and Zierath, 2013; Fujita and Crist, 2018). These biochemical adaptations in response to a variety of physiological cues result in adjustment of cell size and function (Fluck, 2006). For example, upon overload stimuli, muscle mass increases by hypertrophic growth are mediated by an upregulation of global protein production (Bolster et al., 2003). Underlying these changes at the molecular level, activation of mammalian target of rapamycin (mTOR) kinase, which phosphorylates ribosomal S6K kinase and 4E-BP, has been implicated in activating or de-repressing translation (Battaglioni et al., 2022; Csibi et al., 2010). mTOR has also been implicated in regulating protein production by impacting ribosome biogenesis (Chaillou, 2019; Fyfe et al., 2018). Therefore, mTOR signaling appears to exert executive control over skeletal muscle mass by affecting protein translation at several levels of regulation (Coolican et al., 1997; Gugliuzza and Crist, 2022).

Consistent with the idea that changes in nucleolar morphology are linked with changes in protein translational capacity of the cell, in this study, we document profound changes in nucleolar morphology during the differentiation of myogenic cells. We also document that rRNA production and protein translation are markedly upregulated during myogenesis. Congruently, repression of mTOR activity resulted in abrogation of nucleolar morphology changes and inhibition of the myogenic program. Conversely, mTOR gain-of-function experiments resulted in enhanced nucleolar morphology adaptations. Thus, these data implicate mTOR signaling in nucleolar morphology changes that are associated with enhanced ribosome biogenesis and protein translational capacity in differentiating myogenic cells.

The inception of these studies occurred when we were visualizing nucleoli in growing and differentiating muscle cells using well-characterized nucleolar marker proteins. Primarily, we observed a marked difference in the morphology of the nucleoli in that the growing proliferative myogenic progenitor cells (myoblasts; MBs), which contained multiple small nucleoli with a characteristic spherical shape, whereas the multinucleated differentiated cells (myotubes) typically contained one large, often irregularly shaped, nucleolus within the nucleoplasm. A previous study on the effects of creatine on myoblast differentiation also suggested a change in the myoblast nucleoli (Ohira et al., 2011). The morphology of the small spherical nucleoli is consistent with the typical archetypal phase-separated nucleolus described in many previous studies (reviewed in Lafontaine et al., 2021). Conversely, the formation of the large, morphologically irregular shaped nucleoli observed in myotubes and their connection with nucleolar function is less well understood. The nucleolar architecture in myotubes is also less consistent with a typical liquid–liquid phase-separated (LLPS) compartment in that the morphology does not have a spherical droplet shape and might involve a maturation of the biophysical properties of the nucleoli towards a gel or solid-like state. This is a striking change indicating a substantial re-arrangement of genomic structure, given that multiple genetic loci harbor the rRNA repeats from different acrocentric chromosomes (chromosome 12, 15, 16, 18 and 19 in mouse, and chromosome 13, 14, 15, 21 and 22 in humans) (Moss et al., 2019). In view of these novel preliminary observations, we subsequently undertook further analysis of the nucleolar changes occurring in differentiating myogenic cells.

During the differentiation program of myoblasts, many muscle-restricted genes are transcribed and translated in order to form the highly specialized mature muscle cell, which, among many functions, has to contract in response to membrane depolarization and also serve as a major metabolic depot in the control of metabolism. In particular, large amounts of sarcomeric proteins need to be synthesized for the formation of the contractile units, termed sarcomeres. Although transcriptional control of muscle differentiation has been extensively characterized in the last 40 years, the corresponding translational changes underlying this dramatic alteration of the proteome is only beginning to be understood. In view of this, we considered the possibility that the change in nucleolar structure that we observed might be crucial for enhancing ribosome biogenesis necessary to support the vast changes in protein translational capacity required for the execution of the differentiation program.

We thus hypothesized that alteration of nucleolar morphology and function is required for myotube formation. To test this idea, we initially documented the size changes of the nucleoli in differentiating and undifferentiated muscle cells. We chose the C2C12 model culture system, which has long been a well-characterized model for the muscle differentiation program, to study the nucleolar changes. These cells reliably form multinucleated differentiated myotubes (MyoG+ and MyHC+; herein MyHC refers to myosin II family heavy chains) after 4 days in reduced mitogenic differentiation medium (DM), while under the same culture conditions, some cells within the culture become growth arrested and quiescent (reserve cells or RCs; single nucleus, MyoG− and MyHC−). This phenomenon thus allows a direct comparison of the differentiated and undifferentiated phenotype within the same culture with clonogenic (i.e. genetically equivalent) myogenic progenitor cells. To initially mark and visualize the morphology of the nucleoli, we used well-characterized antibodies that recognize nucleolar marker proteins, such as fibrillarin (Fig. 1A; Fig. S1A, also see Movies 1 and 2) and nucleolin (Fig. S1B,C). Given that nucleoli in myotubes and RCs were not necessary located on the same optical plane, we obtained z-stacked images per one field of view and rendered one orthogonally projected image (Fig. S1A–C). This approach allowed the accurate measurement of the size (length) of nucleoli in the myotubes and RCs. As seen in Fig. 1A, in the MyHC+ myotubes, most of the nucleoli were larger in size and less in number per nucleus compared to those of the RCs (magnified images in Fig. 1B). Given that fibrillarin forms clusters of small foci at the DFC phase of the nucleolus, we also stained for nucleolin, which localizes to the GC layer (the outermost region of the tri-partite multiphase nucleolus), to confirm that the nucleoli observed in the myotubes had a continuous GC layer (Fig. 1C; Fig. S1D,E). To clearly identify the nucleolus in the nucleus in two different cellular states in the same field of view, we used MyoG staining as a marker for the differentiating cells to distinguish the multinucleated myotubes from the mono-nucleated RCs (Fig. 1D; Fig. S1F,G). We also extended the analysis to proliferating myoblasts (MBs) in growth medium (GM) (Fig. 1E). The average nucleolar size (average length of nucleoli; mean±s.d.) in MBs, myotubes and RCs were 2.35±1.33 µm (n=225), 4.11±3.18 µm (n=97), and 1.43±0.54 µm (n=480), respectively. Of the three distinguishable populations (MBs, myotubes and RCs) the myotubes exhibited enhanced nucleolar size and less nucleoli compared to both the RCs (P<0.00001) and the MBs (P<0.0001; one-way ANOVA with Tukey HSD post hoc test). To carefully document the nucleolar size distribution in each cell state, we sorted the nucleoli into five categories based on their length (Fig. 1F). These data verify that there is an increased percentage of the large nucleoli (>4.5 µm) in differentiating cells compared to the RCs (Fig. 1F). Given that the C2C12 cell line was originally isolated from clonal colonies derived from mouse leg muscle (Yaffe and Saxel, 1977), we wanted to determine whether the nucleolar changes we documented in C2C12 are also observed in primary mouse skeletal muscle satellite cells. To do this, we utilized myoblasts derived from mouse hindlimb skeletal muscle satellite cells isolated from gastrocnemius and tibialis anterior muscles. Similar alterations in nucleolar morphology and size were observed in primary mouse skeletal myocytes (Fig. 2A,B).

In summary, under differentiation conditions, the size and number of nucleoli were markedly different depending on the cellular differentiation status in both C2C12 cells and primary mouse skeletal muscle myoblasts (satellite cells). In contrast to these observations in myogenic cells, no significant morphological size changes occurred in the nucleoli of fibroblast cell lines (C3H10T1/2 and NIH3T3), when the cells were exposed to DM conditions (P>0.05) (Fig. S2), indicating that the nucleolar changes are a function of the myogenic differentiation program and not a general cellular response to the differentiation conditions.

It has been reported that, generally, there is a positive correlation between the size and biosynthetic activity of the nucleolus. Given that we observed large nucleoli in the differentiated myotubes, and smaller, more abundant ones in the RCs, we postulated that the nucleolar activity might be enhanced in the myotubes compared to the RCs. To test this idea, we took advantage of the fact that the majority (>85%) of RNAs in a cell are rRNAs produced in the nucleolus (Kraus et al., 2019). We therefore used an RNA specific fluorescent dye (SytoRNA Select) to stain accumulated RNAs using a live-cell imaging technique, the rationale being that the majority of RNA signal is generated from rRNA biogenesis, and indeed nucleoli were marked by this RNA dye (Fig. 3A). Remarkably the signal intensity of SytoRNA Select was apparently higher in the myotubes than in the RCs within the same cultures (Fig. 3B). Quantification of signal intensity by 2.5 intensity plot analysis (Fig. 3B) and a line scan (Fig. 3C) of Hoechst 33342 (DNA) and SytoRNA Select signals indicates that signal intensity was considerably higher in the myotubes (green open arrows) than in the undifferentiated RCs (red open arrows). We further converted the intensity of the fluorescence signal derived from SytoRNA Select into a Heat Map (HM) to visualize the overall level of RNAs in the different cellular states, and similar to the observations described above, myotubes had higher signal intensity than RCs (Fig. 3D). To confirm that the enhanced overall RNA production in the myotubes leads to an upregulation of translational capacity, puromycin was added into the medium to label nascently synthesized polypeptides. The cells were then fixed and subjected to immunofluorescence analysis for quantification of active translation by means of visualization of puromycylated peptide accumulation (Fig. 3E). In addition, the translational capacity indicated by puromycin incorporation was linked with the morphology of the nucleoli by nucleolin staining, and for myotube formation by F-actin accumulation (Fig. 3E). Substantially higher accumulation of puromycylated peptides was observed in the myotubes compared to the RCs, confirming a higher translational capacity in tandem with the large increase in rRNA and the changes in nucleolar structure observed. Altogether, global translational activity is higher in differentiating cells and lower in the cells maintaining an undifferentiated state even under the same low mitogen culture conditions. These data collectively infer that there is a dramatic and regulated induction of rRNA biogenesis and protein translation in differentiating myogenic cells that is correlated with nucleolar morphology changes.

Based on the possibility, alluded to above, that a cellular program is invoked to promote nucleolar re-organization and rRNA biogenesis in differentiating myogenic cells, we noted that the mTOR signaling pathway has been previously implicated as an important regulator of nucleolar function in a variety of cellular contexts (reviewed in Battaglioni et al., 2022). Therefore, to test whether mTOR might play a similar role in myocytes, we initially inhibited mTOR activity using the well characterized cell-permeable pharmacological inhibitor, rapamycin, and recorded the morphology of the nucleoli (a working model is outlined in Fig. 4A). We confirmed that components of the canonical mTOR signaling pathway such as ribosomal protein S6 kinase (S6K) and UBF (also known as UBTF), a key transcription factor involved in rRNA gene transcription, are expressed in myogenic cells (Fig. 4B). The levels of two nucleolar proteins, nucleolin and fibrillarin, were upregulated when the cells were exposed to DM and maintained during the differentiation program (after DM2). In addition, we monitored the cellular localization of P-S6K (S389) (Fig. 4C) and total S6K (Fig. 4D) in myotubes and RCs (DM4) by immunofluorescence analysis. We noted that the active, phosphorylated form of S6K (P-S6K) accumulated in the nuclei of both myotubes and RCs, whereas the signal intensity of P-S6K was higher in differentiating cells (MyHC+), as depicted by the HM (Fig. 4E).

Next, we examined whether mTOR inhibition by rapamycin treatment impacted the nucleolar morphology. To assess this, C2C12 cells were maintained in DM for 4 days with or without rapamycin (Fig. 5A). First, in the control condition, in agreement with previous results, the average size of nucleoli in the MyHC+ cells (myotubes) was larger than that of the MyHC− cells (P<0.00001), and the average size of nucleoli in the nucleus of MyHC+ cells treated with rapamycin was almost identical to that of the nucleoli in MyHC− cells (Fig. 5B). Thus, it appears that the nucleolar size changes observed in the differentiation time course are essentially negated by rapamycin treatment. In addition, we observed that 70.0% of the cellular nuclei were in MyHC+ cells (myotubes) in DM, whereas, in the presence of rapamycin, this was reduced to 20.4% (Fig. 5C) indicating that the differentiation program was also inhibited. In addition, in contrast to the control condition where most MyHC+ cells were multinucleated, most if not all MyHC+ cells in rapamycin-treated cultures were mono-nucleated. Thus, rapamycin inhibited both the enlargement of the nucleoli and also subsequent myotube formation. In order to ascertain whether the loss of nucleolar re-organization in rapamycin-treated cells was a secondary effect of inhibiting the onset of differentiation, we added rapamycin to the cultures after 2 days in DM (Fig. 5E). We reasoned that this would allow the cells to initiate differentiation and concomitantly allow us to see whether nucleolar re-organization was still inhibited. At day 2 in DM, C2C12 cells started forming multinucleated myotubes. At day 4, control cells differentiated to form myotubes (MyHC+) (Fig. 5D), and the size of nucleoli were significantly larger than that of RCs (MyHC−) (Fig. 5F). In the presence of rapamycin, multinucleated myotubes were formed but the size of nucleoli was almost identical to that of the RCs indicating that rapamycin inhibited the nucleolar re-organization even in differentiating cells. In agreement with cellular morphology, 80.6% of the cellular nuclei were in MyHC+ cells (myotubes) in DM, whereas in the presence of rapamycin, it was reduced to 58.8% (Fig. 5G). We further substantiated these findings by analysis at an earlier time point (DM2) using MyoG as a differentiation marker (Fig. S3A–C). These data suggest that mTOR signaling has an integral role in nucleolar re-organization and function that is, to some degree, independent of the differentiation program. Based on these data, we propose that nucleolar re-organization is a property of myogenic cells that is not secondary to the differentiation program.

Given that the loss of mTOR kinase activity reported above resulted in the loss of enlarged nucleoli found in differentiating cells, we next used a gain-of-function approach to further address the mechanistic role of mTOR signaling in nucleolar changes. To do this, we employed gain-of-function mutations of the mTOR kinase, which results in hyperactivation of its kinase activity. Exogenous expression of mTOR1 S2215Y, mTOR1 I2500F or empty vector (pcDNA3) was carried out by transient transfection into C2C12 MBs, and transfected cells were marked by expression of a co-transfected GFP expression vector (Fig. 6). Whereas the size of nucleoli in GFP+ and GFP- (non-transfected) cells in the pcDNA3 control group [ratio for GFP+ (n=116) to GFP (n=254)=1.11±1.07; mean±s.d.] were not significantly different, the average area of the nucleoli in mTOR1 S2215Y- and I2500F-expressing cells (GFP+) was, on average 2.87±3.24 [for GFP+ (n=82) versus GFP− (n=226)] and 3.50±3.68 [GFP+ (n=73) versus GFP− (n=407)] times larger than those of non-transfected cells (GFP−) (P<0.00001). Thus, both gain (mTOR1 S2215Y and I2500F) and loss (rapamycin) of function studies indicate a key role of mTOR signaling in nucleolar re-organization in myogenic differentiation.

Collectively, we document novel observations indicating a dramatic mTOR dependent re-organization of nucleoli in differentiating myogenic cells that might underpin substantial changes in ribosomal RNA production and protein translation capacity during muscle cell differentiation.

The nucleolus is a critical phase separated subnuclear compartment that subserves gene expression by functioning as a factory for rRNA transcription and initial steps in ribosome biogenesis. In this study, we document profound changes in nucleolar morphology during the program of myogenic differentiation. Re-organization of nucleolar structure is correlated with a dramatic enhancement of rRNA production and a concomitant increase in protein translation. Further studies employing gain- and loss-of-function strategies revealed that perturbing nucleolar re-organization impedes the myogenic differentiation program. Specifically, we observed that rapamycin treatment, which has been previously characterized as preventing nucleolar size changes by inhibiting mTOR, prevents both nucleolar re-organization and myogenic differentiation. Conversely, gain of function experiments using a hyperactive mTOR mutation promotes nucleolar enlargement in myogenic cells, which is a signature of increased rRNA production and protein translational capacity. Manipulation of mTOR signaling did not alter nucleolar morphology in non-myogenic cells, suggesting that it might be a unique requirement for differentiating muscle cells. These studies document that mTOR signaling regulates an important re-organization of nucleolar structure and function that is required for myogenic differentiation. These novel findings might facilitate new avenues of discovery for therapeutic intervention in muscle pathologies, such as rhabdomyosarcoma and sarcopenia, and in muscle regeneration following injury.

Ribosome biogenesis is one of the most energy consuming cellular processes, as >85% of cellular RNA is composed of rRNAs (Kraus et al., 2019), underscoring why the nucleolus is so fundamental to cellular function. In particular, differentiation into specialized cells requires a massive re-arrangement of gene expression to facilitate the differentiation-specific cellular proteome. Historically, the ribosome was considered as a non-specialized part of the cellular ‘housekeeping’ machinery; however, recent reports highlight the heterogeneity of ribosome composition in differentiated cells. For example, ribosomal protein large 3-like (Rpl3l) is exclusively expressed in striated muscles and its expression is diminished during hypertrophic growth (Chaillou et al., 2016). Therefore, although it has been well established that transcriptional control of mRNA encoding genes is a primary event for both cellular differentiation and functional muscle adaptations, efficient protein synthesis due to selective translation of mRNAs by specialized ribosomes might also prove important, therefore highlighting why a more complete understanding of nucleolar function, rRNA transcription and ribosome biogenesis are imperative. Further characterization of the mTOR signaling pathway and its role in nucleolar organization and activity during myogenesis is thus warranted.

In the model of in vitro muscle differentiation employed here, undifferentiated RCs and differentiating cells (myotubes) are remarkably derived from genetically equivalent clonogenic myoblasts within the same culture. One feature of this bi-partite cell fate decision is that the nucleolar activity and organization of the nucleolus in the two cell states are diametrically opposed. In myotubes, the nucleoli are highly active in rRNA production, morphologically enhanced in size and non-spherical. In RCs, the nucleoli exhibit low levels of rRNA production and are morphologically spherical, small and numerous compared to those in myotubes. In agreement with other reports, there is an established causal relationship between the size of the nucleolus and its activity (Buchwalter and Hetzer, 2017; Neumuller et al., 2013); myotubes have active and larger nucleoli, and RCs conversely have inactive and smaller nucleoli within the same culture conditions. In conditions of hypertrophic growth, a downstream effector of the mTOR kinase, S6K, which is activated by phosphorylation at T389 (P-S6K), accumulates in the myotube nuclei (Bouquier et al., 2020). In agreement with this, we also observed a selective accumulation of P-S6K in the myotube nuclei, suggesting the possibility that mTOR signaling might play an important role in nucleolar re-organization in muscle differentiation and adaptation.

Changes in nucleolar morphology in differentiating myocytes prompted us to consider the underlying biophysical phenomena leading to such alterations. Given that it is well established that the nucleolus is one of the archetypal membrane-less LLPS compartments in the cell, we considered whether the changes observed might be due to modulation of the phase separation properties of the nucleolus. Although the biophysical basis of different phase-separated compartments is quite complex and requires rigorous analysis, we can clearly observe that the typical LLPS spherical nucleoli seen in the undifferentiated RC and myoblast cells gives way to a much larger and non-spherical morphology in the differentiated myotubes, consistent with an alteration in the nucleolar phase properties. This could be due to a transition or ‘maturation’ from a liquid to a gel or solid-like state that has been well documented in the phase separation literature (reviewed in Lafontaine et al., 2021). Moreover, we recently documented that c-Jun targeting to the nucleoli in myogenic cells can lead to a similar change in the nucleolar morphology in myogenic cells (Miyake and McDermott, 2022). The key question now concerns whether the changes observed in morphology are causally linked with alterations in nucleolar function. Preliminary correlative data suggests that this might be the case and this question needs further interrogation. The field of phase separation as it relates to biological systems is still in its infancy but promises to provide definitive tools and knowledge to answer these important questions related to nucleolar morphology and function.

In summary, we have documented substantive re-organization of the nucleolar morphology in differentiating myogenic cells. This property is conserved in primary satellite cells from mouse hindlimb muscle. Nucleolar re-organization is correlated with a dramatic induction of cellular RNA synthesis and protein translation. Characterization of the assembly and maturation of the phase separated properties of the nucleolus will no doubt further our understanding of the global changes in RNA and protein synthesis leading to myogenic differentiation. Moreover, these new insights concerning nucleolar function may spawn new therapeutic approaches to various muscle pathologies and muscle regeneration.

C2C12 myoblasts (CRL-1772), C3H10T1/2 (CCL-226) and NIH/3T3 (CRL-1658) cells were obtained from the American Type Culture Collection. Cells were maintained in growth medium (GM) consisting of Dulbecco's modified Eagle's medium (DMEM; D6429, MilliporeSigma) and supplemented with 10% fetal bovine serum (HyClone) and 1% penicillin-streptomycin (516106, MilliporeSigma) in a moisturized incubator (Model 3120, Thermo Fisher Scientific) at 5% CO₂ and 37°C. Myotube formation was induced by transferring C2C12 cells to differentiation medium (DM; DMEM with 2% FBS) upon reaching 80–90% confluency. Primary mouse skeletal muscle satellite cells were obtained from iXCells Biotechnologies (10MU-033). These cells are derived from mouse skeletal muscle satellite cells isolated from mouse hindlimb gastrocnemius and tibialis anterior muscles.

The cells were thawed according to the manufacturer's protocol and directly plated on glass-bottom dishes (ibidi). The cells were maintained in GM (DMEM+20% FBS) and differentiation was initiated at 70–80% confluency in DM (DMEM+2%FBS) for 4 days.

Anti-P-S389-S6K antibody (B2H9L3: #701064) was obtained from Thermo Fisher Scientific. Anti-nucleolin (D4C7O: #14574 or E5M7K: #87792) antibody was obtained from Cell Signaling technology. Antibodies against MyoG (F5D), MyHC (MF20) and puromycin (PMY-2A4) were obtained from Developmental Studies Hybridoma Bank (The University of Iowa, Department of Biology, Iowa City, IA 52242). Antibodies against S6K (H-9: sc-8418), CKM (C-14: sc-15164), and β-actin (I-19, sc1616) was obtained from Santa Cruz Biotechnology. Antibodies against P-S484-UBF [EPR2725(2): ab182583], Nup153 (QE5:ab24700), and fibrillarin (38F3: ab4566) were obtained from Abcam. For immunoblotting, HRP-conjugated secondary antibodies were obtained from MilliporeSigma (goat anti-rabbit-IgG; #1662408, goat anti-mouse-IgG; #1721011). For immunofluorescence, Alexa Fluor (AF-)488 or 546-conjugated antibodies were obtained from Thermo Fisher Scientific (goat anti-rabbit-IgG conjugated to AF-488: A-11070; goat anti-rabbit-IgG conjugated to AF-546: A-11035; goat anti-mouse-IgG conjugated to AF-546: A-11030).

SYTO RNASelect green fluorescent cell stain (5 mM solution in DMSO: #s32703), CellMask Deep Red Plasma membrane stain (C10046), and Hoechst 33342 solution (20mM: 62249) were obtained from Thermo Fisher Scientific. Rapamycin (#9904) was obtained from Cell Signaling. Puromycin dihydrochloride (PUR333) was obtained from BioShop.

C2C12 cells were seeded on polymer-coated glass-bottom dishes (#81158, ibidi) and fixed with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS: P3813, MilliporeSigma) on ice for 5 mins and for 10 mins at room temperature. To remove PFA, cells were washed three times with PBS and permeabilized with ice-cold 90% Me-OH for 1 min on ice. Cells were washed again three times with PBS and incubated with blocking buffer (5% FBS in PBS) for 2 h at room temperature, and then incubated with the indicated primary antibodies in blocking buffer overnight at 4°C (Table S1). After three washes with PBS to remove the unbound primary antibody, cells were incubated with appropriate Alexa Fluor-conjugated secondary antibody in the blocking buffer at room temperature for 1.5 h. After three washes with PBS, the cells were counter stained with Hoechst 33342 (B2261-25MG, MilliporeSigma) and subjected to confocal fluorescent imaging with a Zeiss Observer Z1 microscope equipped with a Yokogawa CSU-X1 spinning disk equipped with 100× objective lens [α Plan-APOCHROMAT 100× NA1.46 Oil DIC(UV) VIS-IR, 420792-9800, Zeiss]. Images were recorded by AxioCam MRm camera (Zeiss) or an Evolve camera (electron-multiplying CCD camera; Teledyne Photometrics) and processed using Zen 2.5 (blue edition) software (Zeiss).

Cells were seeded on polymer-coated glass-bottom dishes (#81158, Ibidi) and transfected with indicated expression plasmids. Next day, the cell culture medium was replaced with FluoBrite (A1896701, Thermo Fisher Scientific) supplemented with 10% FBS (HyClone) for imaging. During imaging, the cells were maintained in a moisturized environmental chamber (5% CO₂) at 37°C on the confocal fluorescence microscope (Zeiss Observer Z1). For staining of RNAs, SYTO RNASelect (S32703, Thermo Fisher Scientific) was added to the cell culture media to a final concentration of 5µM. For membrane staining, CellMask Deep Red (C10046, Thermo Fisher Scientific) was added according to the manufacturer's protocol. Images were recorded using an Evolve camera (electron-multiplying CCD camera; Teledyne Photometrics) or AxioCam MRm camera (Zeiss) and processed using Zen 2.5 (blue edition) software (Zeiss).

Orthogonal maximum projection images and 3D-rendering images were generated from Z-stack images obtained at 240 nm intervals using Zen 2.5 (blue edition) software (Zeiss). The length of the nucleolus was measured on the orthogonal projection images using Zen 2.5 (blue edition) software (Zeiss). The size of nucleoli was also determined on the threshold orthogonal projection images using Analyze Particles/ImageJ (National Institutes of Health).

Cells were harvested after three washes with PBS. Spun down cells (1000 g for 5 min) were lysed with constant agitation at 4°C for 30 min and total cellular proteins were extracted with NP-40 lysis buffer [1% (v/v) NP-40, 50 mM Tris-HCl pH 8, 150 mM NaCl, and 5 mM EDTA pH 8, 1 mM sodium orthovanadate, 10 mM sodium fluoride, and supplemented with protease inhibitor cocktail (P-8340, MilliporeSigma)]. Protein concentrations were determined by standard Bradford assay. Equivalent amounts of protein were boiled in SDS loading buffer [100 mM Tris-HCl pH 6.5, 200 mM DTT, 4% (w/v) SDS, 3 mM Bromophenol Blue and 15% (w/v) glycerol] for 5 min and then separated by SDS-PAGE with PageRuler prestained protein ladder (#26616, Thermo Fisher Scientific). The proteins separated were transferred to immobilon-FL polyvinylidene difluoride (PVDF) membrane (IPFL00010, MilliporeSigma) from the SDS-PAGE gels according to the manufacturer's instructions. The membrane was blocked with blocking buffer (5% skim milk in PBS) prior to incubation with primary antibody at 4°C overnight (Table S1) with agitation. After three washes with PBS, appropriate HRP-conjugated secondary antibody was added and incubated for 2 h at room temperature. After washing three times with PBS, the protein–antibody immunocomplexes were detected by a chemical reaction with Enhanced Chemiluminescence (ECL) substrate (WBLUC0500, MilliporeSigma) and recorded using an iBright CL1500 imaging system (Thermo Fisher Scientific). The images were processed using iBright software (Thermo Fisher Scientific).

Puromycin was added to the medium (final concentration to 0.5 µM) and maintained for 15 mins in the cell culture incubator. The cells were washed three times with PBS and fixed and permeabilized and subjected to immunofluorescence analysis with anti-puromycin antibody (PMY-2A4).

For ectopic protein expression, cells were transfected with a mixture of Lipofectamine 2000 (#11668027, Thermo Fisher Scientific) and indicated mammalian expression DNA constructs diluted with Opti-MEM (#31985062, Thermo Fisher Scientific) according to the manufacturer's protocol.

Mammalian expression constructs for pcDNA3-FLAG-mTOR-S2215Y was Addgene #69013 and pcDNA3-FLAG-mTOR-I2500F was Addgene #69014) deposited by David Sabatini (Grabiner et al., 2014).

Length of nucleoli of MBs, myotubes, and RCs on the orthogonal projection images was measured using Zen software. Size (area) of nucleoli of cells on the orthogonal projection images was measured using ImageJ. The average size (area) and the s.d. were calculated and average fluorescence intensity of the size of nucleoli in each condition was analyzed by one-way ANOVA and Tukey's HSD pairwise comparisons within the ANOVA data.

We acknowledge the York University microscopy facility.

The peer review history is available online at https://journals.biologists.com/jcs/lookup/doi/10.1242/jcs.260496.reviewer-comments.pdf