Abstract
The pharyngeal muscles of the nasal, oral, and laryngeal pharynxes are required for swallowing. Pharyngeal muscles are preferentially affected in some muscular dystrophies yet spared in others. Muscle stem cells, called satellite cells, may be critical factors in the development of pharyngeal muscle disorders; however, very little is known about pharyngeal satellite cells (PSC) and their role in pharyngeal muscles. We show that PSC are distinct from the commonly studied hindlimb satellite cells both transcriptionally and biologically. Under basal conditions PSC proliferate, progress through myogenesis, and fuse with pharyngeal myofibers. Furthermore, PSC exhibit biologic differences dependent on anatomic location in the pharynx. Importantly, PSC are required to maintain myofiber size and myonuclear number in pharyngeal myofibers. Together, these results demonstrate that PSC are critical for pharyngeal muscle maintenance and suggest that satellite cell impairment could contribute to pharyngeal muscle pathology associated with various muscular dystrophies and aging. Stem Cells 2015;33:3581–3595
Pharyngeal muscles are essential for swallowing and are preferentially affected in some muscular dystrophies yet spared in others. Muscle stem cells, called satellite cells, may be critical factors in the development of pharyngeal muscle disorders; however, very little is known about pharyngeal satellite cells (PSC) and their role in pharyngeal muscles. Here we characterize satellite cells of pharyngeal muscles demonstrating that pharyngeal satellite cells undergo myogenesis under basal conditions. Additionally, PSC are required for maintenance of both pharyngeal myofiber size and myonuclear number, suggesting that satellite cell impairment could contribute to pharyngeal myopathies.
Introduction
Muscular dystrophies are a group of degenerative muscle diseases that impair different subsets of skeletal muscles depending on the specific type of muscular dystrophy [1-8]. One muscle group differentially affected in muscular dystrophies is found in the pharynx. Pharyngeal muscles are a vital group of seven muscles involved in swallowing. They are non-somitic in origin, arising from the cranial mesoderm of the third and fourth pharyngeal arches in vertebrates [9, 10]. These muscles surround the nasal, oral, and laryngeal pharynxes and include the palatopharyngeus, stylopharyngeus, salpingopharyngeus, superior and middle pharyngeal constrictors, cricopharyngeus, and thyropharyngeus muscles [11-16]. One type of muscular dystrophy in which pharyngeal muscles are pathologically affected is oculopharyngeal muscular dystrophy (OPMD), a late onset autosomal dominant disease caused by a polyalanine expansion in the N-terminal domain of the ubiquitously expressed polyadenylate-binding protein nuclear 1 (PABPN1) protein [6, 17, 18]. Interestingly, the associated muscle stem cells of pharyngeal muscles are also pathologically affected in OPMD patients [19].
Muscle stem cells, called satellite cells, are a heterogeneous cell population that is responsible for repair of muscle tissue [20]. In hindlimb skeletal muscle, satellite cells are mitotically quiescent under basal conditions [20]. When muscle tissue is damaged or injured, satellite cells proliferate, differentiate, migrate, adhere and fuse to each other or existing myofibers to form multi-nucleated myofibers while a subset of satellite cells undergo self-renewal to maintain a quiescent stem cell population [20]. Considering that pharyngeal satellite cells are altered in OPMD and that satellite cells in other skeletal muscles are thought to play a role in the pathology of muscle diseases such as Duchenne muscular dystrophy [21], congenital muscular dystrophy 1A [22], Emery-Dreifuss muscular dystrophy [23], and facioscapulohumeral muscular dystrophy [24, 25] we addressed whether pharyngeal satellite cells may have unique biological properties that make them susceptible to disease-inducing conditions.
In this study, we analyzed the biologic properties of pharyngeal satellite cells (PSC) and their contribution to pharyngeal muscle maintenance. We characterized PSC in vivo in a region-dependent manner: analyzing PSC of the palatophayrngeus muscle (nasal and oral pharynx) along with the cricopharyngeal and thyropharyngeal muscles (laryngopharynx). Somite-derived satellite cells from hindlimb muscles were used for comparison. We found that PSC are distinct from hindlimb satellite cells both transcriptionally and biologically. PSC undergo constitutive myogenesis and, unlike hindlimb satellite cells [26-30], are required to maintain myofiber size and myonuclear number in pharyngeal myofibers. Our findings provide new insights into the biology of PSC and pharyngeal muscles that may be important in understanding why certain muscular dystrophies target muscles of the pharynx.
Materials and Methods
Mice
Adult male mice, between 2 and 4 months of age, were used unless noted otherwise. C57BL/6 mice were purchased from Charles River Laboratories (Wilmington, MA, USA; www.criver.com). Myf5nLacZ/+ (Myf5 nLacZ) and Pax7CreERTM/CreERTM (Pax7CreERTM) mice were obtained from S. Tajbakhsh [31] and C. Keller [32], respectively. Duchenne muscular dystrophy model mice containing a dystrophin-deficient allele with a splice site mutation in exon 23, C57BL/10ScSn-Dmdmdx/J (Mdx) [33], were purchased from Jackson Laboratories (Bar Harbor, ME; www.jax.org). Rosa26-CAG-tdTomato [34] and Rosa26-DTA176 mice [35] were also purchased from Jackson Laboratories. Homozygous Pax7CreERTM/CreERTM male mice were crossed with either homozygous RosaDTA176/DTA176 (DTA) females to obtain Pax7CreERTM/+; RosaDTA176/+ (DTA-Pax7CreERTM) mice for satellite cell ablation experiments, or with homozygous RosatdTomato/tdTomato (tdTom) to obtain Pax7CreERTM/+; RosatdTomato/+ (tdTom-Pax7CreERTM) mice to fluorescently label myogenic cells after tamoxifen treatment. Genomic recombination and removal of floxed stop sequences were induced in male DTA-Pax7CreERTM and tdTom-Pax7CreERTM mice at 8 weeks-of-age. Tamoxifen, 1 mg (Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com) per 10 grams body weight, was injected intraperitoneally once daily for 5 days. Flow cytometry was used to determine the recombination efficiency in both DTA-Pax7CreERTM and tdTom-Pax7CreERTM mice. Experiments were performed in accordance with approved guidelines and ethical approval from Emory University's Institutional Animal Care and Use Committee and in compliance with the National Institutes of Health.
Dissection of Pharyngeal Tissue
CO2 asphyxiation was used to euthanize mice immediately before tissue collection. Pharyngeal tissue dissection was performed as described previously [16]. Histologic samples included pharyngeal tissue extending from the soft palate caudally to the cranial aspects of the trachea and esophagus. The larynx and trachea were excluded from pharyngeal samples collected for isolation of myogenic cells.
Flow Cytometry and Fluorescence Activated Cell Sorting
For analysis via flow cytometry, mononucleated cells were isolated from pharyngeal and hindlimb (gastrocnemius and quadriceps) muscles as described previously [36, 37]. Briefly, pharyngeal and hindlimb muscles were minced and digested in Dulbecco's modified Eagle's medium (DMEM) (Cellgro Mediatech, Manassas, VA; www.cellgro.com) containing 1 mg/ml pronase (Calbiochem, Billerica, MA; www.emdmillipore.com), 25 mM HEPES at 37°C for 45 minutes or 1 hour, respectively. Cellular preps were applied to Percoll (GE Healthcare, Pittsburgh, PA; www.gelifesciences.com) gradients of 20% and 60% for enrichment of myogenic cells and removal of red blood cells [38]. Digested muscles were washed with DMEM and mononucleated cells collected using 100 μm Steriflip filtration systems (Milipore, Billerica, MA; www.emdmillipore.com) before antibody labeling.
For analysis and collection via fluorescence-activated cell sorting (FACS), pharyngeal and hindlimb (gastrocnemius and quadriceps) muscles were minced and digested in Ham's F10 media (Hyclone, Pittsburgh, PA; www.gelifesciences.com) containing 500 units/ml collagenase II (Gibco, Grand Island, NY; www.lifetechnologies.com) and 10% fetal bovine serum (FBS) at 37°C while shaken at 65 rpm for 90 minutes. Digested muscles were then rinsed with Ham's F10 media containing 10% FBS, 100 U/ml penicillin G, and 100 μg/ml streptomycin (P/S), followed by a second digestion using 100 units/ml collagenase II, 1 unit/ml dispase (Gibco, Grand Island, NY; www.lifetechnologies.com) in Ham's F10 media containing 10% FBS, P/S under the same conditions for 30 minutes. Digested muscles were washed with 0.1 M Dulbecco's phosphate-buffered saline (PBS), pH 7.3 (Gibco, Grand Island, NY; www.lifetechnologies.com) and mononucleated cells collected using 100 μm Steri-Flip filtration systems (Milipore, Billerica, MA; www.emdmillipore.com).
Isolated cells were resuspended in PBS containing 1% bovine serum albumin (BSA) for antibody labeling. Dead cells were identified using 5 μg/ml propidium iodide (PI). Myogenic cells, identified as PI−/Sca1−/CD31−/CD45−/α7-integrin+ (Lin−α7-integrin+) [39] were isolated and collected using a FACSAria II (Becton-Dickinson, Franklin Lakes, NJ; www.bd.com) at the Emory University School of Medicine Core Facility for Flow Cytometry. Analyses of flow cytometry data were performed using FlowJo (version 9.5.2). Isolated myogenic cells were then processed for in vitro cultures, immunofluorescent staining, or RNA extraction. Primary antibodies included rat anti-CD31-Phycoerythrin (PE) (0.5 μg/ml; eBioscience, San Diego, CA; www.ebioscience.com) and rat CD45-PE (0.5 μg/ml; BD Pharmingen, San Jose, CA; m.bdbiosciences.com), rat Sca-1-PE-Cy7 (0.05 μg/ml; BD Pharmingen, San Jose, CA; m.bdbiosciences.com), rat α7-integrin-AlexaFluor649 (1 μg/ml; AbLabs, Vancouver, British Columbia), and appropriate rat isotype control antibodies (BD Pharmingen, San Jose, CA; m.bdbiosciences.com).
Immunohistochemistry/Immunofluorescence
All muscle tissues were cryofrozen in Tissue Freezing Medium (Triangle Biomedical Sciences, Durham, NC; www.trianglebiomedical.com) and stored at −80°C. Tissue sections of 10 μm thickness were collected every 80 or 200 μm using a Leica CM1850 cryostat to analyze pharyngeal and hindlimb muscles, respectively. Tissue sections from Myf5-nLacZ mice were fixed in 4% paraformaldehyde, 0.1 M NaPi (pH 7.2), 0.5% glutaraldehyde, followed by a PBS wash, and then incubated at 37°C with 1 mg/ml X-gal in dimethylformamide for 12–18 hours to identify β-galactosidase activity. Peripheral and centrally localized β-gal positive nuclei were quantified using ImageJ 1.43u.
For immunostaining of tissue sections for dystrophin and bromo-2′-deoxyuridine (BrdU), the M.O.M. Kit (Vector Laboratories Inc., Burlingame, CA; www.vectorlabs.com) was used to block endogenous Fc receptor binding sites followed by a 1 hour incubation with 5% goat serum, 5% donkey serum, 0.5% BSA, 0.20% Triton-X 100 in PBS (blocking buffer). Sections were then labeled with blocking buffer containing primary antibodies or appropriate isotype controls overnight at 4°C. Successive washes in 0.25% Tween 20 in PBS (PBS-T) were followed by another 60-minute incubation with appropriate secondary antibodies at room temperature. Sections were then washed with PBS-T and nuclei labeled with 4′,6-diamidino-2-phenylindole (DAPI) (1 μg/ml) in PBS. Primary antibodies included rat anti-BrdU (2 μg/ml, clone BU1/75 (ICR1), Accurate Chemical and Scientific), and mouse anti-dystrophin (13.5 μg/ml, clone MANDYS8, Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com). Immunostaining for BrdU was performed on post-dystrophin immunostained sections as follows. Sections were fixed with 2% paraformaldehyde in PBS for 10 minutes, followed by a PBS wash, 2 minute chilled acetone treatment, and another PBS wash. Subsequently, sections were treated with 1 N HCl in PBS at 45°C for 20 minutes and immediately neutralized with a 0.1-M sodium tetraborate/boric acid buffer, pH 8.5, for 8 minutes at room temperature. Additional tissue blocking was performed with 0.3 M glycine in PBS for 30 minutes and 24 μg/ml AffiniPure goat α-mouse IgG F(ab′)2 (Jackson ImmunoResearch, West Grove, PA; www.jacksonimmuno.com)/ml PBS-T at room temperature for 1 hour. Sections were then immunostained for BrdU following the standard immunostaining protocol described above.
For immunostaining of Pax7 and laminin, tissue sections from hindlimb (tibialis anterior) and pharyngeal muscles were fixed in freshly prepared 4% paraformaldehyde (Electron microscopy sciences, Hatfield, PA; www.electronmicroscopysciences.com), followed by PBS washes. Sections were heated in citrate buffer (10 mM sodium citrate, 0.05% Tween-20, pH 6.0) for antigen retrieval with a pressure cooker (Cuisinart model CPC-600) for 6 minutes. Before immunostaining of tissue sections, the M.O.M. Kit (Vector Laboratories Inc., Burlingame, CA; www.vectorlabs.com) was used to block endogenous Fc receptor binding sites followed by 1-hour incubation with 10% goat serum, 10% donkey serum, 0.5% BSA, 1% Triton-X 100 in PBS. Sections were then labeled with mouse anti-Pax7 (3.8 μg/ml, Developmental Hybridoma Studies Bank, Iowa City, IA; dshb.biology.uiowa.edu) and rabbit anti-laminin (2 μg/ml, Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com) antibodies or IgG isotype controls overnight at 4°C. Successive washes in PBS-T were followed by incubating in 0.3% hydrogen peroxide for 10 minutes to block endogenous peroxidase. Sections were incubated for 20 minutes with biotinylated goat-anti-mouse F(ab′)2 IgG fragments (1 μg/ml, Jackson ImmunoResearch, West Grove, PA; www.jacksonimmuno.com) for Pax7 imaging. A Tyramide Signal Amplification (TSA; Perkin Elmer, Waltham, MA; www.perkinelmer.com) Green kit was used for subsequent steps. Sections were incubated in TNB (Tris-NaCl blocking buffer) for 20 minutes followed by incubation with HRP-conjugated streptavidin diluted 1:200 in TNB for 20 minutes. Finally, signals were amplified by TSA Green diluted 1:300 in amplification buffer for 5 minutes. After further washes with PBS-T, sections were incubated with AF594-donkey anti-rabbit F(ab′)2 IgG fragments for 40 minutes for laminin imaging. Sections were then washed using PBS-T and nuclei labeled with DAPI in PBS. Sections were mounted with Vectashield (Vector lab, Burlingame, CA; www.vectorlabs.com).
Immunofluorescent staining for Pax7 was performed on Lin−α7-integrin+ and Lin−α7-integrin− cells collected via FACS. Cytospun cells were incubated with the M.O.M. kit and blocking buffer sequentially for one-hour intervals. Cells were incubated with mouse anti-Pax7 (7.5 μg/ml, Developmental Studies Hybridoma Bank, Iowa City, IA; dshb.biology.uiowa.edu) at RT for 1 hour, followed by PBS-T washes, and futher labeled using biotinylated goat-anti-mouse F(ab′)2 IgG fragments. Pax7 staining was visualized using the TSA Green kit, as described above. Cells were then washed with PBS-T and nuclei labeled with DAPI. Samples were then mounted with Vectashield for imaging.
Myofiber cross-sectional areas of 929–1,505 myofibers were quantified on hematoxylin (Thermo Scientific, Waltham, MA; www.thermofisher.com) and eosin (Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com) stained sections using ImageJ 1.43u. For all tissue section studies, four, five, or six representative sections imaged at 200× magnification were analyzed from muscles of the nasal, oral, and laryngeal pharynx, respectively, while four sections were analyzed per tibialis anterior muscle. Images were acquired at 23°C using an Axioplan microscope with a 0.5 NA 20× Plan-Neofluar objective (Carl Zeiss MicroImaging, Inc., Oberkochen, Germany; www.zeiss.com) and charge-coupled device camera (Carl Zeiss MicroImaging, Inc., Oberkochen, Germany; www.zeiss.com) with Scion Image 1.63 (Scion Corp., NIH, USA). Photoshop CS4 (Adobe, San Jose, CA; www.adobe.com) was used to globally process all images for contrast, size, and brightness. Data were blinded for analysis.
Single Myofiber Isolation and Imaging
Single myofibers were isolated, as described previously [40] from pharyngeal muscles of mice 2–5, 12, or 18 months-of-age. Briefly, pharyngeal muscles were dissected and digested with gentle agitation in 4.5 mg/ml glucose, 100 U/ml penicillin G, 100 μg/ml streptomycin, 25 mM HEPES in DMEM containing either 400 or 800 U/ml collagenase type I (Worthington), respectively, at 37°C for 90 minutes. Single myofibers were washed three times with collagenase-free DMEM before manual extraction onto clean 100-mm plates. Myofibers were then individually transferred onto 10% growth factor-reduced Matrigel (BD Biosciences, San Jose, CA; m.bdbiosciences.com) coated 24-well plates, and centrifuged at 1,100g to enhance myofiber adherence. Myofibers were then fixed with 2% formaldehyde in PBS and stained with DAPI. Pharyngeal muscles from 18-month-old mice required digestion with 1600 U collagenase type I/ml DMEM media. Single myofiber images were visualized at 23°C using an Axiovert 200M microscope with a 0.3 NA 10× Plan-Neofluar objective (Carl Zeiss MicroImaging, Inc., Oberkochen, Germany; www.zeiss.com). Images were captured with a QImaging camera and OpenLab 5.5.2 (Improvision, Waltham, MA; www.perkinelmer.com) software. Myofiber diameter, length, and nuclear number were quantified using ImageJ 1.43u. Photoshop CS4 (Adobe, San Jose, CA; www.adobe.com) was used to globally process all images for contrast and brightness. Between 929 and 1505 myofibers were pooled and analyzed from five mice per age group for each muscle.
Evans Blue Assay
C57BL/6 and Mdx mice received an intraperitoneal injection of 1% Evans Blue Dye (Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com) suspended in sterile PBS at a volume of 10 μl/gram body weight. Mice were euthanized via CO2 asphyxiation 24 hours post-injection and immediately before dissection and collection of tibialis anterior and pharyngeal muscles for cryosectioning. Tissues were sectioned at a thickness of 10 μm and analyzed for the presence of Evans Blue fluorescence within myofibers. Images were acquired at 23°C using an Axioplan microscope with a 0.8 NA 25× Plan-Neofluar objective (Carl Zeiss MicroImaging, Inc., Oberkochen, Germany; www.zeiss.com) and charge-coupled device camera (Carl Zeiss MicroImaging, Inc., Oberkochen, Germany; www.zeiss.com) with Scion Image 1.63 (Scion Corp., NIH, USA). Photoshop CS4 (Adobe, San Jose, CA; www.adobe.com) was used to globally process all images for contrast, size, and brightness.
In Vivo BrdU Assays
To compare the proliferative and fusogenic abilities of pharyngeal and hindlimb satellite cells in vivo, 5-BrdU assays were performed. Three-month-old male C57BL/6 mice were injected with 10 μg BrdU (Sigma-Aldrich, St. Louis, MO; www.sigmaaldrich.com)/gram body weight intraperitoneally every 12 hours. To assess proliferation, mice were injected over a 48-hour period. To assess fusion, injections were given for 7 days followed by 7 days of 0.8% BrdU in 2% sucrose water.
Proliferation Assay
Pharyngeal and gastrocnemius muscles of 3–5 mice were collected 48 hours post-initial BrdU injection. Mononucleated cells were isolated from pooled muscles, selected with Percoll gradients, immunostained, fixed with formalin, and analyzed using flow cytometry, as described above. Proliferating myogenic cells were identified as BrdU+, Sca1−/CD31−/CD45−/α7-integrin+. BrdU immunostaining was performed using the FITC BrdU Flow Kit (1:200, BD Pharmingen, San Jose, CA; m.bdbiosciences.com). As a positive control for proliferating cells, muscle injuries were performed in anesthetized mice by injecting 40 μl of 1.2% BaCl2 into gastrocnemius muscles as described previously [41] 2 days before collection.
Fusion Assay
Pharyngeal and tibialis anterior muscles were collected 14 days post-initial BrdU injection as described for tissue sectioning. Representative tissue sections were immunostained to detect dystrophin and BrdU as described above. The number of intrafiber BrdU+ myonuclei/100 myofibers was quantified using ImageJ 1.43u. Samples were blinded for analysis.
Microarray
Myogenic cells from pharyngeal and hindlimb (gastrocnemius and quadriceps) muscles were isolated, sorted using FACS, and collected from 10 to 30 mice per experiment. Samples were sent to the Emory University Integrated Genomics Core facility for total RNA isolation using Qiagen miRNEAsy kit with on-column DNAse treatment followed by one round of amplification using NuGEN's WT-Ovation Pico amplification kit. Analysis of genomic gene expression was performed using an Illumina Mouse WG-6 v2.0 Expression BeadChip. Data were extracted using the Illumina HiScan Scanner and iScan control software. Illumina Genome Studio 2011.1 software suite was used to normalize probe level intensity data with background correction using manifest MouseWG-6_V2_0_R1_11278593_ A.txt. Detection p-values were calculated as the proportion of negative control probes with expression greater than the regular probe in question using Partek Genome Studio. The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [42] and are accessible through GEO Series accession number GSE69418 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE69418).
Data were further analyzed using both MetaCore Genego (https://portal.genego.com; Thomson Reuters) and Gene Set Enrichment Analysis (www.broadinstitute.org/gsea/index.jsp; Broad Institute).
Real-Time Polymerase Chain Reaction
Myogenic cells from pharyngeal and hindlimb (gastrocnemius and quadriceps) muscles were isolated, sorted using FACS, and collected from 10 to 30 mice per experiment. Total RNA was isolated using the PicoPure RNA Isolation Kit (Applied Biosystems, Grand Island, NY; www.lifetechnologies.com) from samples of 150,000–200,000 pooled cells each. cDNA was generated from total RNA via a reverse transcriptase reaction using M-MLV reverse transcriptase (Invitrogen, Grand Island, NY; www.lifetechnologies.com) and random hexamer primers. cDNA was then amplified using the SYBR Select Master Mix reagent (Applied Biosystems, Grand Island, NY; www.lifetechnologies.com) and 2.5 μM of each primer. All RNA samples were tested for DNA contamination by polymerase chain reaction (PCR). Primer sequences were: MyoD (F: 5′- GCCCGCGCTCCAACTGCTCTGAT-3′ and R: 5′-CCTACGGTGGTGCGCCCTCTGC-3′); Pax7 (F: 5′-CACCCCGGGGACAGAGGAAGAT-3′ and R: 5′-GAGAGGGCGGGGTACAAGGAAGAC-3′). All other primers were purchased from Qiagen's RT2 qualitative PCR (qPCR) Primer Assay library. Real-time PCR (RT-PCR) reactions were performed and analyzed with a StepOnePlus Real Time PCR System (Applied Biosystems, Grand Island, NY; www.lifetechnologies.com), using GAPDH or 18S as an internal control. Fold change of gene expression was determined using the ΔΔCt method [43]. Three to four independent experiments were performed and analyzed in duplicate.
Clonal Expansion Assay
Conditioned media (CM), collected from either primary hindlimb or pharyngeal muscle cell cultures, contained 20% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, 5 ng/ml basic fibroblast growth factor (bFGF) in Ham's F10 media. Any contaminating cells were removed through sterile syringe filtration with a 0.20 μm filter (Corning, Corning, NY; www.corning.com). Mononucleated cell suspensions obtained from pharyngeal and hindlimb (gastrocnemius and quadriceps) muscles of C57BL/6 mice were then subjected to FACS, as described previously. Sorted myogenic cells were plated at clonal densities of 100–250 cells/100 mm collagen-coated plates and grown in CM diluted 1:1 with Ham's F10 media containg 20% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 5 ng/ml bFGF at 37°C for 8 days before 2% paraformaldehyde fixation. Cells were then counterstained with hematoxylin (Thermo Scientific). Bright field images were aquired at 8× magnification using a Leica MZFL III microscope (Leica, Inc., Buffalo Grove, IL; www.leica-microsystems.com) with a Planapo 1.0× objective (Leica, Inc., Buffalo Grove, IL; www.leica-microsystems.com). Images were captured with a QImaging Retiga Exi Fast 1394 camera and QCapture Suite version 2.70 (Quantitative Imaging Corp., Surrey, British Columbia; www.qimaging.com) software. Analyses of cell number per colony and number of clonal colonies were performed on 264 clones. Photoshop CS4 (Adobe, San Jose, CA; www.adobe.com) was used to globally process all images for contrast and brightness.
Statistical Analyses
Data were analyzed for statistical significance using GraphPad Prism version 5 for Macintosh (GraphPad Software, San Diego, CA; graphpad.com). For all statistical tests, a 0.05 level of confidence was considered statistically significant. When comparing two groups, data were analyzed by unpaired Student's t test. To determine significance among multiple groups, data were analyzed using one-way analysis of variance with Bonferroni's posttest. Nonparametric data were analyzed using the Kruskal-Wallis test with Dunn's posttest to identify statistical differences between sample distributions.
Results
Pharyngeal muscles contain larger numbers of activated satellite cells than hindlimb muscles in the absence of overt injury
To date, satellite cells of adult mouse pharyngeal muscles, which arise from the third and fourth pharyngeal arches during development, have not been studied in depth, unlike satellite cells of muscles arising from the first and second pharyngeal arches [44-46]. Therefore, to gain insights into PSC we initially analyzed their numbers in pharyngeal muscles. The pharynx can be subdivided into three distinct regions (nasal, oral, and laryngeal) (Fig. 1A). Each region and their associated muscles vary in both location and function. We recently reported that muscles within the nasal, oral, and laryngeal regions of the pharynx are differentially affected by aging and disease [16]. Therefore, all histologic studies were analyzed based on pharyngeal localization. To analyze satellite cell numbers throughout each region of the pharynx, we used a mouse model that expresses nuclear-localized β-galactosidase (β-gal) under the promoter of the myogenic transcription factor Myf5 (Myf5-nls-LacZ), which is commonly used to mark both quiescent and activated satellite cells within muscle tissues [31, 47, 48]. X-gal staining of hindlimb and pharyngeal muscle sections from 3 to 5 month old mice (Fig. 1B) revealed significantly increased numbers of β-gal+ cells (Fig. 1C), located at the periphery of myofibers throughout the oral and laryngeal pharynxes compared with hindlimb muscle. To differentiate between quiescent and activated satellite cells, we immunostained pharyngeal muscle sections for paired box protein 7 (Pax7), a marker of quiescent satellite cells (Fig. 1E) [49]. Pax7+ satellite cell numbers were significantly decreased in the nasopharynx while increased in the laryngopharynx compared with hindlimb muscle (Fig. 1F). The increased numbers of Pax7+ satellite cells in the murine laryngopharynx are consistent with recent immunohistologic studies of human cricopharyngeal muscles [50]. Most strikingly, our Myf5-β-gal data demonstrate that the majority of satellite cells are activated in pharyngeal muscles.
![Pharyngeal muscles contain a larger number of activated satellite cells than hindlimb muscle. (A): Schematic representation of murine pharyngeal regions shows NP, OP, and LP (modified from Randolph et al. [16]). (B): Hindlimb (tibialis anterior) and pharyngeal muscles were collected at 9 weeks of age from Myf5-nlacZ mice, sectioned, and incubated with X-gal to identify β-gal+ nuclei (blue). Representative muscle sections from hindlimb and oral pharynx (palatopharyngeus) are shown. Two types of β-gal+ nuclei were observed: peripherally located to myofibers (arrows) or centrally located within myofibers (∧). (C): Each pharyngeal region contained increased numbers of peripherally located β-gal+ nuclei (satellite cells) versus hindlimb muscle. (D): Increased numbers of centrally localized β-gal+ myonuclei in myofibers were also observed in each pharyngeal region relative to hindlimb muscle. Data in (C) and (D) represent the mean ± SEM. *, p < 0.05. n = 4 mice. (E, F): Representative muscle sections from 8 to 9 week old C57BL/6 mice immunostained for Pax7 (green) and laminin (red). Pax7+ cells colocalizing with 4′,6-diamidino-2-phenylindole and laminin were identified (E) and quantified (F). Scale bar = 100μm. Data represent the mean ± SEM. *, p < 0.05. n = 4 mice. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; LP, laryngopharynx; NP, nasal pharynx; OP, oral pharynx; TA, tibialis anterior muscle.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/stmcls/33/12/10.1002_stem.2098/2/m_stmcls_33_12_3581_f1.jpeg?Expires=1712757208&Signature=yJG5FLaswnBj3OFiW33rcisjgTwR-qEeiKfECo9gqfhVs~T5EjgXwEQhYZeim8hIN9ZYzeRiGESYoFWw5tqAP~hIteHp7Ee97kkktddk-CKXbPdFEN6tGvkCIihOqxRHae6ybCM0Bap4DFHIsM29~eRh-zihxfND2pJYHkF3c2sUtJaGCNy2aNPqecMcFbaCHyAxZVkJU3NdxW-uj2cg-dldndM~jCLV0q0X4iXe9DPIZiCIEypqN~-B-GQxvIA1tjcs1cO1B35K3jXyw2Ykres4S3EVdhP3bG6BwCG-pZMVSxtMaUWr~N20jtLVjwAouGvow06mqElhf4BaENr0lw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Pharyngeal muscles contain a larger number of activated satellite cells than hindlimb muscle. (A): Schematic representation of murine pharyngeal regions shows NP, OP, and LP (modified from Randolph et al. [16]). (B): Hindlimb (tibialis anterior) and pharyngeal muscles were collected at 9 weeks of age from Myf5-nlacZ mice, sectioned, and incubated with X-gal to identify β-gal+ nuclei (blue). Representative muscle sections from hindlimb and oral pharynx (palatopharyngeus) are shown. Two types of β-gal+ nuclei were observed: peripherally located to myofibers (arrows) or centrally located within myofibers (∧). (C): Each pharyngeal region contained increased numbers of peripherally located β-gal+ nuclei (satellite cells) versus hindlimb muscle. (D): Increased numbers of centrally localized β-gal+ myonuclei in myofibers were also observed in each pharyngeal region relative to hindlimb muscle. Data in (C) and (D) represent the mean ± SEM. *, p < 0.05. n = 4 mice. (E, F): Representative muscle sections from 8 to 9 week old C57BL/6 mice immunostained for Pax7 (green) and laminin (red). Pax7+ cells colocalizing with 4′,6-diamidino-2-phenylindole and laminin were identified (E) and quantified (F). Scale bar = 100μm. Data represent the mean ± SEM. *, p < 0.05. n = 4 mice. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; LP, laryngopharynx; NP, nasal pharynx; OP, oral pharynx; TA, tibialis anterior muscle.
We then analyzed whether muscle damage was present in pharyngeal muscles, which would induce a regenerative response resulting in increased numbers of activated satellite cells. We observed no signs of myofiber degeneration in sections despite the fact that Myf5-nls-LacZ pharyngeal myofibers contained a high incidence of centrally located β-gal+ myonuclei compared with hindlimb muscles (Fig. 1D), suggestive of recent satellite cell fusion with myofibers [47]. To confirm that no overt myofiber damage was present, we injected mice with Evans Blue and muscle sections were analyzed for fluorescence. Damaged myofibers were identified by the presence of Evans Blue dye within the myofiber, as seen in mdx mice, which are characterized by damaged myofiber membranes due to the loss of dystrophin (Supporting Information Fig. 1). However, no Evans Blue positive myofibers were observed in any of the wild-type pharyngeal muscles examined (Supporting Information Fig. 1). Therefore, the increase in activated satellite cell numbers in pharyngeal muscles could not be attributed to myofiber damage, suggesting that alternative mechanisms are responsible for the increased satellite cell numbers and centrally located myonuclei of pharyngeal muscles.
Identification of Pharyngeal Satellite Cells
To gain further insights into the biology of pharyngeal satellite cells, we initially confirmed which molecular markers would reliably identify these cells in vivo. Satellite cells in different muscles can be identified using multiple molecular markers [39, 51-57], but such markers have not been validated for PSC. Hindlimb satellite cells (LSC) have previously been identified as cells expressing α7-integrin that are not of endothelial, hematopoietic, or fibro-adipogenic progenitor lineages (CD31−CD45−Sca1−) [39]. To validate whether these markers also identified PSC, pharyngeal muscles were enzymatically dissociated to yield mononucleated cells. Analysis by flow cytometry revealed a distinct sub-population of lineage-negative (CD31−CD45−Sca1−), α7-integrin expressing (Lin−α7-integrin+) cells present within pharyngeal tissue (Fig. 2A). To confirm the myogenic identity of pharyngeal Lin−α7-integrin+ cells, two different methods were used to confirm expression of Pax7 [49] (Fig. 2B–2D). Over 90% of FACS-sorted pharyngeal Lin−α7-integrin+ cells were positive for Pax7 via immunofluorescence in vitro (Fig. 2B, 2C). To confirm these results, a tamoxifen-inducible Cre/LoxP system was used. Mice expressing a tamoxifen-inducible Cre recombinase under the endogenous Pax7 promoter (Pax7CreERTM) [32], were crossed with mice containing the floxed fluorescent reporter Rosa26-tdTomato allele (tdTom) [34] to visualize satellite cells in vivo. Over 90% of pharyngeal Lin−α7-integrin+ cells from tamoxifen treated tdTom-Pax7CreERTM heterozygotes expressed tdTomato (Fig. 2D), further confirming that Lin−α7-integrin+ labeling identifies Pax7+ satellite cells from pharyngeal muscles.
Identification of pharyngeal satellite cells using established cellular markers for hindlimb satellite cells. (A): Flow cytometry gating to identify and sort satellite cells isolated from pharyngeal muscle. Pharyngeal satellite cells were identified as α7-Integrin+ cells not derived from endothelial (CD31), hematopoietic (CD45), or fibro-adipogenic progenitor (Sca1) lineages (Lin−α7-Integrin+). (B, C): Lin−α7-Integrin+ cells were sorted and Pax7 expression visualized (B) and quantitated (C) in vitro using immunofluorescence. n = 509 pooled cells from 2 experiments of 10 mice each. (D): Approximately 90% of pharyngeal Lin−α7-Integrin+ cells were tdTomato+ following tamoxifen-treatment of Pax7CreERTM/RosatdTomato heterozyotes. n = 2 experiments of 2–3 mice pooled. Scale bar = 20 μm. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; FITC, fluorescein isothiocyanate; PI, propidium iodide.
Pharyngeal Satellite Cells Proliferate in the Absence of Induced Injury
Satellite cells of some craniofacial muscles, such as extraocular muscles, proliferate at a basal level in the absence of injury, in contrast to the quiescent phenotype of LSC [58-61]. Considering that overt muscle injury did not explain the large number of PSC or the centrally located myonuclei throughout the pharynx, we tested whether PSC were proliferating in the absence of induced injury. Bromodeoxyuridine (BrdU) assays were performed in both young and mature C57BL/6 mice to label proliferating cells. Mice received BrdU injections every twelve hours for two days (Fig. 3A). On day three, hindlimb and pharyngeal muscles were collected, mononucleated cells isolated, and myogenic (Lin−α7-integrin+) cells analyzed via flow cytometry for evidence of BrdU labeling (Fig. 3B). At 2–5 months of age, we observed a trend towards increased satellite cell proliferation in the pharynx (Fig. 3C), and by 12 months of age, the percentage of proliferating satellite cells was significantly increased by approximately 30-fold in pharyngeal muscle compared with hindlimb (Fig. 3C). In sharp contrast to quiescent hindlimb satellite cells, myogenic cells from pharyngeal muscle demonstrated robust proliferation in the absence of induced injury suggesting that basal satellite cell proliferation is a characteristic of pharyngeal muscles.
Pharyngeal satellite cells proliferate and fuse with pharyngeal myofibers in the absence of induced injury. (A): Schematic of bromo-2′-deoxyuridine (BrdU) treatment protocol. (B): PI−CD31−CD45−Sca1−α7-Integrin+ (Lin−α7-Integrin+) satellite cells were identified using flow cytometry (left column) and analyzed to determine the percentage of proliferating satellite cells (BrdU+) in each tissue (right column). n = 3 mice pooled. (C): Quantification of BrdU+ Lin−α7-Integrin+ cells demonstrated a significantly larger proliferating population of satellite cells in pharyngeal versus hindlimb (gastrocnemius) muscles. Data represent the mean ± SEM. **, p < 0.0001, n = 3 experiments, 3–5 mice per experiment. (D): Lin−α7-Integrin+ cells from either hindlimb (gastrocnemius/quadriceps) or pharyngeal muscles were sorted, plated at clonal densities and cultured for 8 days. Cultures were then fixed and stained with hematoxylin for quantification of cell number per clone. The number of large clones (>300 cells) was increased three-fold in pharyngeal versus hindlimb cultures. n = 264 clones from two experiments using 10 mice each. (E): Schematic of BrdU treatment protocol. (F): Muscle sections were immunostained for dystrophin (red) and BrdU (green). BrdU+ nuclei contained within a dystrophin+ myofiber outline represent satellite cells that recently proliferated and fused into myofibers. Scale bar = 20 μm. (G): Satellite cell fusion was quantified as the number of intrafiber BrdU+ nuclei per 100 myofibers. Satellite cell fusion occurred with higher frequency in pharyngeal muscles compared with hindlimb muscles. *, p < 0.05, n = 4 mice. Abbreviations: BrdU, bromo-2′-deoxyuridine; L, hindlimb; P, pharynx; TA, tibialis anterior.
The basal proliferation of pharyngeal satellite cells could be in response to signals from the pharyngeal niche and/or an intrinsic property of PSC. To examine whether the proliferative SC phenotype in pharyngeal muscles was intrinsic to PSC, satellite cells were isolated and sorted using Fluorescence Activated Cell Sorting (FACS) from both hindlimb and pharyngeal muscles. Sorted myogenic cells were plated at clonal densities and grown for 8 days in the absence of the pharyngeal muscle niche. Cultures were then fixed and stained with hematoxylin to identify nuclei. We identified highly proliferative cells as clones that produced colonies numbering more than 300 cells. While the majority of pharyngeal clones were similar in size to those of the hindlimb, 17% of pharyngeal colonies contained more than 300 cells, compared with only 6% of hindlimb colonies (Fig. 3D; Supporting Information Fig. 2). These data suggest that a highly proliferative subpopulation of PSC exists in pharyngeal muscles under basal conditions, thus indicating that proliferation is an intrinsic property of some PSC.
Proliferating Pharyngeal Satellite Cells Progress through the Myogenic Lineage
Following activation and proliferation, satellite cells typically progress through the myogenic lineage, contributing new myonuclei to muscle fibers [20]. However, impaired differentiation [46] or cell death [62] can prevent myogenic lineage progression. The increased incidence of centrally localized myonuclei in pharyngeal muscles (Fig. 1D) suggested that satellite cell fusion with myofibers does occur. To definitively test whether proliferating PSC progressed through the myogenic lineage in vivo, proliferating cells were labeled with BrdU over a two-week period to allow for fusion of BrdU+ cells into myofbers (Fig. 3E). During myogenic differentiation, satellite cells exit the cell cycle and are post-mitotic. Therefore, any BrdU+ myonucleus located within a myofiber would indicate recent fusion of a proliferating satellite cell. We analyzed pharyngeal muscle sections for the presence of BrdU+ myonuclei within dystrophin immunostained myofibers (Fig. 3F). The number of intra-fiber BrdU+ nuclei was significantly elevated in myofibers of the oral and laryngeal pharynxes relative to those within hindlimb muscle (Fig. 3G). This result indicates that the proliferative satellite cells of pharyngeal muscles progress through the myogenic lineage and contribute new myonuclei to pharyngeal myofibers.
Pharyngeal and Hindlimb Satellite Cells Are Transcriptionally Distinct
To address potential molecular mechanisms involved in the constitutive myogenesis of PSC, we analyzed steady-state RNA levels of sorted Lin−α7-integrin+ cells using both quantitative RT-PCR (qRT-PCR) and microarray. qRT-PCR was used to analyze transcript levels of several transcription factors involved in craniofacial muscle development and myogenesis. Pax7, Myf5, MyoD, Myogenin, and Pitx2 transcripts were all decreased in PSC (Fig. 4A) relative to LSC, Pax3 was not detected in PSC, while Tcf21 was expressed only in PSC (data not shown). These data demonstrate a unique transcription profile of myogenic regulatory factors in PSC. Principal component analysis of microarray data revealed distinct expression profiles in pharyngeal versus hindlimb satellite cells (Fig. 4B). Furthermore, the steady-state levels of 964 pharyngeal transcripts were differentially expressed by ≥1.5 fold. These included 478 up-regulated and 486 down-regulated transcripts (Supporting Information Table 1; Fig. 4C). Consistent with our in vivo data, analysis of PSC transcripts revealed an enrichment of genes involved in regulation of cell proliferation (GO process: 0042127; Fig. 4D). Interestingly, we found increased expression of several cytokine and chemokine transcripts in PSC, including Lif, Ccl2, Ccl7, IL6, and IL6Ra, all of which are involved in satellite cell proliferation and/or differentiation [63-69]. qRT-PCR analysis confirmed increased expression of Lif, Ccl2, and IL6 in PSC relative to LSC (Fig. 4E). Together, these data demonstrate that PSC have a distinct molecular signature from those of hindlimb muscles, which could contribute to their increased proliferation and myogenic lineage progression.
Comparative transcriptome analyses reveal pharyngeal and hindlimb satellite cells are distinct. Gene-expression analyses of FACS sorted pharyngeal and hindlimb satellite cells as determined by qRT-PCR (A, E) and microarray (B–D). (A): Selected regulatory transcripts involved in myogenesis were analyzed via qRT-PCR using RNA isolated from FACS sorted pharyngeal satellite cell (PSC) and LSC. Data represent the mean fold-change of transcript steady-state levels ± SEM. n = 3 experiments each containing 150,000–200,000 satellite cells pooled from 10 to 30 mice. *, p<0.05. (B): Principal component analysis (PCA) of pharyngeal satellite cells (PSC, red dots) versus hindlimb (gastrocnemius/quadriceps) satellite cells (LSC, blue dots). PCA coordinates (PC1, 29.2%; PC2, 22.6%; and PC3, 18.4%) revealed a total data variation of 70.2%. n = 3 experiments each containing 200,000 satellite cells pooled from 10 to 30 mice. (C): Heat maps comparing the levels of the top 50 transcripts either up- or down-regulated in PSC relative to LSC. Steady-state RNA levels are represented with a linear color scale ranging from dark red (enriched) to dark blue (depleted). Transcripts marked with red asterisks were validated by qRT-PCR. (D): Gene ontology (GO) process networks enriched in PSC generated with MetaCore Genego software. GO networks related to cell proliferation are highlighted in red. (E): qRT-PCR was used to validate microarray data of selected cytokine/chemokine transcripts that were enriched in PSC relative to LSC. Data represent the mean fold-change of transcript steady-state levels ± SEM. n = 3 hindlimb and 4–5 pharyngeal experiments each containing 1,500,000–200,000 satellite cells pooled from 10 to 30 mice. *, p<0.05. Abbreviations: IL, interleukin; LIF, leukemia inhibitory factor; LSC, hindlimb satellite cell; PSC, pharyngeal satellite cell; Ccl, chemokine (C-C motif) ligand.
Pharyngeal Satellite Cells Are Required to Maintain Myonuclear Number in Pharyngeal Muscles
Our data show that PSC are transcriptionally distinct from LSC, highly proliferative, and undergo myogenesis under basal conditions. Considering the continuous myonuclear addition occurring in pharyngeal muscles, myonuclear numbers within pharyngeal myofibers would be predicted to increase over time. To test this hypothesis, we isolated individual myofibers from both hindlimb and pharyngeal muscles, stained myofibers with DAPI to visualize nuclei (Fig. 5A), and compared myonuclear numbers using three independent metrics from mice at various ages. No increase in myonuclear number was observed across the studied age range (Fig. 5B–5D). These data, in conjunction with the presence of a proliferative satellite cell population, suggest that pharyngeal muscles undergo active myonuclear turnover in the absence of induced injury. To directly test whether myonuclear loss was occurring in pharyngeal myofibers, we crossed the Pax7CreERTM mouse with mice containing the floxed truncated diptheria toxin A-176 allele in the Rosa26 locus (DTA-Pax7CreERTM) [35] to induce ablation of Pax7+ satellite cells in vivo (Fig. 5E). When Pax7+ cells were ablated following tamoxifen-induced expression of DTA, a 93% loss of Lin−α7-integrin+ cells was observed (Fig. 5F; Supporting Information Fig. 3). We hypothesized that removing approximately 93% of all satellite cells from pharyngeal muscles would severely impair myonuclear addition and, if myonuclear turnover were occurring in pharyngeal muscle, loss of myonuclei would be evident over time. Therefore, we collected pharyngeal muscles from DTA-Pax7CreERTM mice 4 months following treatment with either vehicle or tamoxifen (Fig. 5E). Muscle sections were labeled for dystrophin to identify pharyngeal myofibers and counterstained with DAPI to visualize myonuclei contained within myofibers. Myonuclear numbers were significantly decreased, compared with vehicle, in myofibers of the nasal pharynx but not the oral or laryngeal pharynxes (Fig. 5G). Myonuclear loss following satellite cell ablation suggests that Pax7+ satellite cells are required to maintain myonuclear numbers in nasal pharyngeal muscles.
Myonuclear turnover occurs in pharyngeal muscle under basal conditions. (A): Merged 4',6-diamidino-2-phenylindole (DAPI) and phase contrast images of a representative myofiber isolated from pharyngeal muscles. Scale bar = 50μm. (B–D): Quantification of various pharyngeal myonuclear parameters indicated no change in myonuclear numbers with age. n = 26–35 fibers per timepoint. (E): Schematic of satellite cell-specific ablation in Pax7CreERTM/RosaDTA−176 heterozygous mice. (F): Pharyngeal Lin−α7-Integrin+ cells were ablated following tamoxifen-treatment of Pax7CreERTM/RosaDTA−176 (DTA/Pax7CreERTM) heterozygotes. Ablation efficiencies for pharyngeal Lin−α7-Integrin+ cells ranged from 87%–97%. n = 8 experiments of 2–3 mice pooled. (G): Quantification of DAPI+ nuclei contained within dystrophin+ myofiber outlines revealed myonuclear loss within satellite cell-ablated muscles of the nasal pharynx. *, p<0.05, n = 3–4 mice per condition. Abbreviations: TM, tamoxifen; V, vehicle.
Loss of Pharyngeal Satellite Cells Results in Muscle Atrophy but not Fibrosis
Muscle fibrosis can occur when satellite cells are impaired, such as in aged muscles or in satellite cell-ablated hindlimb muscles following overload stress [70, 71]. Therefore, we tested whether fibrosis also occurred in the pharyngeal muscles of our satellite cell-ablated mice. Hematoxylin and eosin staining of nasal pharyngeal muscle sections revealed no obvious increase in extracellular matrix (ECM) deposition in tamoxifen-treated mice compared with vehicle (Supporting Information Fig. 4A). These data were confirmed using fluorescently labeled wheat germ agglutinin to visualize N-acetyl-d-glucosamine glycosylated ECM proteins (Supporting Information Fig. 4B). No significant difference in wheat germ agglutin staining was observed between corresponding regions of the nasal pharynx of vehicle versus tamoxifen treated mice (Supporting Information Fig. 4C). Oral and laryngeal regions showed no differences between vehicle versus tamoxifen treated cohorts as well (data not shown). These data demonstrate that satellite cell ablation does not cause fibrosis in pharyngeal muscles.
Interestingly, myonuclear loss has been associated with muscle atrophy resulting from various stimuli [40, 62, 72, 73]. Therefore, we tested whether satellite cell ablation, which resulted in myonuclear loss, also led to pharyngeal muscle atrophy. Pharyngeal muscle sections from tamoxifen treated DTA-Pax7CreERTM mice (Fig. 5E) were stained with hematoxylin/eosin and myofiber cross-sectional areas from each pharyngeal region were measured (Fig. 6A). Myofiber size was significantly decreased in satellite cell-ablated muscles of the nasal pharynx versus DTA-Pax7CreERTM mice injected with vehicle (Fig. 6B). No change in myofiber size was observed in the oral or laryngeal pharynx (Fig. 6B). Importantly, muscle atrophy correlated with myonuclear loss in the nasal pharynx. These data suggest that pharyngeal satellite cells are required to maintain myofiber size in pharyngeal muscles of the nasal pharynx.
Pharyngeal satellite cells are required to prevent muscle atrophy in the nasal and laryngeal pharynxes. DTA/Pax7CreERTM mice received either vehicle or tamoxifen injections as described in Fig. 5E with pharyngeal muscles collected 4-months post-treatment. (A): Histologic sections of vehicle and tamoxifen treated mice. Scale bar = 50μm. (B): Frequency distribution plots of myofiber cross-sectional areas from the naso-, oro-, and laryngopharyngeal regions are shown. Myofiber size of nasal pharyneal muscles decreased following satellite cell ablation. n = 929–1505 myofibers, three mice per condition.
Discussion
Here we propose a novel role for satellite cells in the maintenance of pharyngeal muscles (Fig. 7). PSC are transcriptionally distinct satellite cells that contribute new myonuclei to pharyngeal myofibers through constitutive myogenesis in the absence of injury. PSC are required to maintain both pharyngeal myonuclear numbers and myofiber size, counteracting the results of active myonuclear turnover within pharyngeal muscles. As discussed below, these results give new insights into why pharyngeal muscles are affected in some muscular dystophies yet spared in others.
Model of basal pharyngeal satellite cell biology and maintenance of myofiber size. Pharyngeal satellite cells (red) proliferate, progress through myogenesis, and contribute myonuclei (black) to pharyngeal myofibers under basal conditions. The continual contribution of new myonuclei (light green) to pharyngeal myofibers counteracts the basal myonuclear loss (grey), preventing both loss of myonuclear numbers and myofiber size. Pharyngeal satellite cell impairment reduces myonuclear addition to pharyngeal myofibers resulting in both myonuclear loss and decreased myofiber size.
Our finding that PSC proliferate under basal conditions in the absence of overt injury is in sharp contrast to the basal quiescence of satellite cells in adult hindlimb skeletal muscles [20]. Although previous groups have identified proliferative satellite cells in adult extraocular muscles (EOM), only approximately 1% or less of EOM satellite cells proliferate under basal conditions compared with the approximately 30% of PSC that proliferate in adult pharyngeal muscle [58, 59, 61]. However, a recent in vivo study of EOM satellite cells proposed that the proliferative phenotype observed was, in part, due to a specific highly proliferative subpopulation [74]. Our in vitro clonal expansion data also suggest that a subpopulation of PSC is contributing to the robust proliferative phenotype observed in pharyngeal muscle. These results are in agreement with the fact that satellite cells are a heterogenous population with differences among cells in myogenic phenotypes [46, 75-77]. PSC are also distinct from hindlimb satellite cells at the transcriptional level. Besides expressing a unique profile of myogenic regulatory factors, PSC also express unique chemokine/cytokines compared with LSC. Interleukin-6, LIF, and monocyte chemoattractant protein-1 (MCP-1; CCL2), which were highly expressed in PSC, all stimulate myoblast proliferation through autocrine and paracrine signaling [63-65, 67-69]. This cytokine expression may drive the proliferative phenotype of PSC. Potentially, myonuclear loss may also contribute to the proliferative phenotype of PSC by directly or indirectly activating proliferative signaling pathways. For example, if nuclear material were expelled into the extracellular milleu during myonuclear turnover, nuclear molecules could induce signaling pathways, like the high-mobility group box 1 (HMGB1) and receptor for advanced glycation end-products (RAGE) pathway, that activate satellite cell proliferation [78, 79]. The potential relationship between myonuclear turnover and pharyngeal satellite cell activation is an intriguing hypothesis that warrants investigation. Further studies are needed to determine the signaling pathways and mechanisms involved in PSC activation and proliferation in vivo.
Proliferating PSC also progress through the myogenic lineage and contribute new myonuclei to pharyngeal muscle fibers under basal conditions. To date, only one other group of muscles, the global and orbital EOM, exhibit a similar basal state of myonuclear addition [58, 59, 61, 80, 81]. Of note, we found a correlation between satellite cell number and the incidence of satellite cell fusion with myofibers in a region-dependent manner in pharyngeal muscles. These results suggest that the increased levels of cell fusion in specific pharyngeal regions are likely due to the presence of the large resident satellite cell population rather than an increased cellular propensity for fusion. Interestingly, fusion of PSC into myofibers over a period of 16 months did not result in myonuclear accretion within pharyngeal myofibers. These findings suggested that a basal level of myonuclear loss or turnover occurs in pharyngeal muscles. To examine myonuclear turnover more directly, we ablated satellite cells in vivo and examined myonuclear numbers using pharyngeal muscle sections. In contrast to recent studies where satellite cell ablation did not alter myonuclear numbers in non-injured hindlimb muscles [27, 28], decreased numbers of myonuclei were observed in satellite cell-ablated muscles of the nasal pharynx. Together these data suggest that maintenance of myonuclear numbers is dependent on basal satellite cell myogenesis in pharyngeal muscles, and for the first time demonstrate myonuclear loss occurring in the absence of associated injury, disease or aging in skeletal muscle.
We also analyzed the functional outcomes of PSC ablation. A decrease in myofiber size was associated with myonuclear loss in nasopharyngeal muscles, suggesting that constitutive myonuclear addition is required to maintain nasopharyngeal muscle size. Interestingly, the muscles of the oro- and laryngopharynx possess larger numbers of activated satellite cells compared with the nasal pharynx. Despite loss of over 90% of PSC from these muscles, satellite cell numbers may have remained above a critical threshold leading to maintenance of myonuclear numbers in these regions. Another functional role of satellite cells in hindlimb muscles involves maintenance of the extracellular mileu [71]. Interestingly, we found no fibrotic changes associated with satellite cell ablation in pharyngeal muscles, suggesting that PSC may play less of a role in extracellular matrix maintenance than recently proposed for LSC. In hindlimb skeletal muscles, satellite cells are not required for maintenance of muscle size or myonuclear number in sedentary mice [26-30]. However, our data provide evidence that satellite cells are required for maintenance of myofiber size in pharyngeal muscles and suggest that the rate of myonuclear turnover within a muscle determines whether or not the associated satellite cells are required for muscle maintenance.
To date, aging [73], muscle atrophy [72], and muscular dystrophy [82] are the major processes associated with decreases in myonuclear number, yet the molecular mechanisms that regulate myonuclear turnover remain unclear. Several groups have suggested myonuclear loss occurs in hindlimb and extraocular muscles by apoptosis using terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assays or caspase staining [62, 83, 84]. However TUNEL staining identifies DNA damage in the context of apoptosis but also necrosis and autolytic cell death pathways [85]. Alternate mechanisms such as nuclear autophagy [86, 87] or even myonuclear extrusion [88] could also contribute to myonuclear turnover. Further studies are needed to elucidate the mechanisms by which myonuclear loss occurs in pharyngeal muscles as well as why myonuclear loss occurs in these muscles.
We recently reported that both aging and disease differentially affect pharyngeal muscles dependent on their regional location within the pharynx [16]. Here, we show PSC also display differential phenotypes depending on the associated muscle and its location within the pharynx. For example, the palatopharyngeus muscle extends between the nasal and oral pharynxes, yet differs in both structure and function depending on its regional location within the pharynx [16]. Interestingly, in the nasal pharynx, the palatopharyngeus consistently demonstrated the lowest number of satellite cells (both quiescent and activated) and fusion of all the pharyngeal regions examined. Additionally, the nasal palatopharyngeus was sensitive to satellite cell ablation as both myonuclear loss and decreased myofiber size occurred. This was in contrast to the palatopharyngeus of the oral pharynx, which had larger numbers of activated PSC and fusion yet demonstrated a resistance to functional changes associated with satellite cell ablation. Additionally, PSC number and fusion were greatly increased in the cricopharyngeal and thyropharyngeal muscles of the laryngeal pharynx compared with other regions of the pharynx. Together, these data suggest that differences in regional location and physiological function of pharyngeal muscles also contribute to the unique properties of PSC.
Conclusion
Our results may provide insights into why pharyngeal muscles are affected in some muscular dystrophies yet spared in others. Considering the requirement of PSC in maintenance of pharyngeal myofiber size, mutations or conditions that adversely affect PSC numbers, activation, proliferation, differentiation, or fusion could negatively impact pharyngeal muscle function. We recently observed a significant decrease in lick rates, indicative of swallowing difficulties, in a mouse model of OPMD that correlated with decreases in central myonuclear localization in pharyngeal muscles, suggesting myonuclear addition was negatively affected [16]. Additionally, some disease-causing mutations could enhance myonuclear turnover in pharyngeal muscles to such an extent that PSC could no longer adequately supply enough myonuclei to maintain homeostasis. Aging also adversely affects pharyngeal function, resulting in impaired swallowing in 11%–16% of the elderly population [89, 90]. Impairment of satellite cell function with aging [91] could contribute to this age-related increase in swallowing disorders. Further studies addressing the effects of disease and aging on both pharyngeal satellite cells and pharyngeal myonuclear turnover are warranted and could lead to new therapeutics for individuals suffering from pharyngeal myopathies.
Acknowledgments
We thank Drs. J.P. Canner, T. Caspary, A. English, M. Johns III, T. Lever, and R. Price for training and/or helpful discussions; E. Andreas for providing the schematic of pharyngeal anatomy; K. Carroll and N. Pescatore for technical assistance; Emory University's Integrated Genomics Core and School of Medicine Core Facility for Flow Cytometry for experimental assistance; and Drs. S. Tajbahksh and C. Keller for the generous gifts of the Myf5nLacZ/+ (Myf5 nLacZ) and Pax7CreERTM/CreERTM (Pax7Cre) mice, respectively. This work was supported by NIH Grants AR061987 (G.K.P.) and DC012225 (M.E.R.).
Author Contributions
M.E.R.: conception and design, financial support, collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript; B.L.P., H.-J.C., and K.E.V.: collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript; Y.V.: collection and assembly of data, data analysis and interpretation, final approval of manuscript; G.K.P.: conception and design, financial support, administrative support, provision of study materials, data analysis and interpretation, manuscript writing, final approval of manuscript.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.
