Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

mRNP granule proteins Fmrp and Dcp1a differentially regulate mRNP complexes to contribute to control of muscle stem cell quiescence and activation

mRNP granule proteins Fmrp and Dcp1a differentially regulate mRNP complexes to contribute to... Background: During skeletal muscle regeneration, satellite stem cells use distinct pathways to repair damaged myofibers or to self-renew by returning to quiescence. Cellular/mitotic quiescence employs mechanisms that promote a poised or primed state, including altered RNA turnover and translational repression. Here, we investigate the role of mRNP granule proteins Fragile X Mental Retardation Protein (Fmrp) and Decapping protein 1a (Dcp1a) in muscle stem cell quiescence and differentiation. Methods: Using isolated single muscle fibers from adult mice, we established differential enrichment of mRNP granule proteins including Fmrp and Dcp1a in muscle stem cells vs. myofibers. We investigated muscle tissue homeostasis in adult Fmr1-/- mice, analyzing myofiber cross-sectional area in vivo and satellite cell proliferation ex vivo. We explored the molecular mechanisms of Dcp1a and Fmrp function in quiescence, proliferation and differentiation in a C2C12 culture model. Here, we used polysome profiling, imaging and RNA/protein expression analysis to establish the abundance and assembly status of mRNP granule proteins in different cellular states, and the phenotype of knockdown cells. * Correspondence: jdhawan@ccmb.res.in; jdhawan@instem.res.in Nainita Roy, Swetha Sundar, Malini Pillai and Farah Patell-Socha contributed equally to this work. Institute for Stem Cell Science and Regenerative Medicine, Bangalore, India Centre for Cellular and Molecular Biology, Hyderabad, India Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Roy et al. Skeletal Muscle (2021) 11:18 Page 2 of 28 Results: Quiescent muscle satellite cells are enriched for puncta containing the translational repressor Fmrp, but -/- not the mRNA decay factor Dcp1a. MuSC isolated from Fmr1 mice exhibit defective proliferation, and mature myofibers show reduced cross-sectional area, suggesting a role for Fmrp in muscle homeostasis. Expression and organization of Fmrp and Dcp1a varies during primary MuSC activation on myofibers, with Fmrp puncta prominent in quiescence, but Dcp1a puncta appearing during activation/proliferation. This reciprocal expression of Fmrp and Dcp1a puncta is recapitulated in a C2C12 culture model of quiescence and activation: consistent with its role as a translational repressor, Fmrp is enriched in non-translating mRNP complexes abundant in quiescent myoblasts; Dcp1a puncta are lost in quiescence, suggesting stabilized and repressed transcripts. The function of each protein differs during proliferation; whereas Fmrp knockdown led to decreased proliferation and lower cyclin expression, Dcp1a knockdown led to increased cell proliferation and higher cyclin expression. However, knockdown of either Fmrp or Dcp1a led to compromised differentiation. We also observed cross-regulation of decay versus storage mRNP granules; knockdown of Fmrp enhances accumulation of Dcp1a puncta, whereas knockdown of Dcp1a leads to increased Fmrp in puncta. Conclusions: Taken together, our results provide evidence that the balance of mRNA turnover versus utilization is specific for distinct cellular states. Keywords: Quiescence, mRNP granule, Translational control, mRNA decay, Skeletal muscle, Myoblast, G0, Fmrp, Dcp1a, Fmr1 knockout, Muscle stem cell Background and stress granules. P-bodies are dynamic structures that During skeletal muscle regeneration, the resident muscle are enriched in proteins involved in mRNA decay (such stem cells, called satellite cells (MuSc), use distinct path- as Dcp1a, Edc4, Edc3, Lsm1-7 complex) [13, 14]. In con- ways to either enter myogenic differentiation to restore trast, stress granules may form in response to stress and functional tissue, or self-renew by returning to mitotic contain stalled translation initiation complexes (contain- quiescence to replenish the stem cell compartment. We ing Fmrp, eIF-4E, eIF-4G, Pabp, Tia-1/TiaR), and occa- have previously reported transcriptional and epigenetic sionally, 40S ribosomal subunits [15]. P-bodies and mechanisms that control the choice between these irre- stress granules share many proteins and interact with versible and reversible cell cycle arrests [1–3]. In par- each other, precluding unambiguous classification based ticular, quiescence is regulated by mechanisms that [15] on the presence or absence of individual compo- promote a poised or primed state, compatible with re- nents, and prompting the use of the inclusive nomencla- entry into the cell cycle [4]. The view of quiescence (G0) ture of “mRNP granules”. Recent reports of mRNP as an actively managed poised state, rather than an inert granule proteins in MuSC suggest a role for translational default state, is supported by several findings [5], which control and use of stored mRNA in regulation of quies- show that two major programs (the cell cycle and myo- cence and activation [16, 17] and in regeneration [18]. genesis) are held in abeyance by diverse mechanisms [6– The flux of transcripts between mRNP granules is as- 8]. In addition to transcriptional and epigenetic silencing sociated with cell state transitions, and altered aggrega- in G0, quiescent cells also exhibit translational repres- tion status is reported in neuromuscular diseases, the sion [9], but remain capable of rapid remobilization of best-studied example being fragile X syndrome (FXS). In pre-existing transcripts onto polysomes during cell cycle this devastating neuro-developmental disorder, loss of activation. the fragile X mental retardation protein (Fmrp) leads to Upon export from the nucleus, newly synthesized a spectrum of autistic features characterized by cognitive mRNAs are either rapidly assembled onto polysomes for and behavioral deficits [19]. The location of Fmrp in immediate translation or held in a non-translating com- cytoplasmic granules and its molecular function as a re- partment, bound by a variety of RNA-binding proteins pressor of activity-dependent protein translation in that control mRNA transport, localization, decay, and axons [20] suggest mechanisms by which signal- translational efficiency. RNA-binding proteins dynamic- dependent protein synthesis is required for higher-level ally coalesce, along with mature mRNAs and miRNAs, brain function [21]. Despite the nearly ubiquitous ex- into mRNP granules [10, 11]. Depending on the cellular pression of Fmrp [17], little is known of its specific role context and lineage, several kinds of mRNP granules in non-neuronal tissues, including development and re- exist [10] that, due to their distinct composition, may generation of skeletal muscle. Fmrp has been reported function differently to regulate mRNA utilization [12]. to be downregulated during muscle differentiation [22] Among mRNP granules, the best studied are P-bodies and is detected in quiescent MuSC, where it regulates Roy et al. Skeletal Muscle (2021) 11:18 Page 3 of 28 -/- MuSC function via control of the myogenic determinant using Fmr1 mice, suggesting its involvement in Myf5 mRNA [17, 23]. homeostatic and regenerative control in muscle, beyond Increasing evidence points to a role for post- its established role in neuronal function. During primary transcriptional control in quiescent cells, including qui- MuSC activation on myofibers, we found that Fmrp and escent adult MuSC. Studies in yeast and cultured fibro- Dcp1a show reciprocal expression and organization, with blasts showed that Fmrp and the related Fxr1 are Fmrp puncta prominent in quiescence, but Dcp1a important for entry into G0 [24]. In myoblasts, the entry puncta appearing during activation/proliferation. We into mitotic quiescence is associated with an induction also explored the muscle cell-intrinsic functions of Fmrp of genes encoding mRNP granule components, such as as distinct from neurological effects manifested in vivo, tristetraprolin (TTP) [25, 26], primarily involved in AU- using a cultured myoblast model of quiescence. Our re- rich element (ARE)-mediated decay, that are also re- sults suggest the existence of distinct mRNP complexes quired for MuSC regenerative function [27]. A pioneer- in different cellular states (proliferation, quiescence, and ing report by Crist et.al., [17] showed that quiescent differentiation). Specifically, whereas translational re- MuSC sequester transcripts of Myf5 in an untranslated pressive complexes containing Fmrp predominate in G0, fraction, and remobilize them onto polysomes during re- we report an enrichment of nonsense-mediated mRNA activation. Further, a general repression of protein syn- decay complexes containing the mRNA-decapping en- thesis by phosphorylation of the translation initiation zyme 1A (Dcp1a) in proliferating myoblasts, suggesting factor eIF2α is essential for maintenance of the quiescent that post-transcriptional regulatory complexes may be state [9]. However, there is little information available remodeled depending on cellular context. Functional on the composition and function of heterogeneous analysis using mRNA knockdown indicates that Fmrp mRNP granules that might regulate the quiescent state and Dcp1a play opposing roles in myogenic proliferation per se, and in particular, the relative roles of transla- and quiescence; Fmrp sustains proliferative potential, tional repression and mRNA decay in the entry into qui- whereas Dcp1a functions to restrain proliferation of escence. This balance may be important in the context myoblasts. Intriguingly, these opposing functions cross- of the global suppression of macromolecular synthesis, regulate, such that knockdown of Fmrp leads to in- in G0 cells [16]. For example, key regulators of protein creased Dcp1a expression and assembly into puncta, and synthesis such as mTOR control awakening of quiescent reciprocally, Dcp1a knockdown myoblasts show in- MuSC [28], but the coupling of mRNA utilization to creased Fmrp expression and assembly into puncta. metabolic activation has not been extensively explored However, unlike their opposing roles in proliferation, in MuSC. knockdown of either Dcp1a or Fmrp led to compro- MuSC function is intimately linked to the ability to mised myogenic differentiation. Taken together, our enter and exit quiescence. Whereas both differentiation study shows the importance of the balance between and quiescence are mitotically inactive states, muscle translational repression and mRNA decay in the regula- terminal differentiation is irreversible and requires pref- tion of quiescence, and indicates a role for distinct erential transcription and translation of tissue-specific mRNP granule proteins in regulating this equilibrium. proteins that comprise and control the specialized sarco- meric cytoskeleton [16]. By contrast, quiescence is re- Materials and methods versible and is characterized by a broad suppression of Single myofiber isolation and analysis the differentiation program and increased expression of Animal experiments were carried out in accordance with the MuSC-specific transcription factor Pax7, which are CPCSEA guidelines of the Govt. of India as approved by reversed during cell cycle activation, along with re- the Institutional Animal Ethics Committee of InStem induction of determination factors MyoD and Myf5 [26]. and CCMB, or in accordance with British law under pro- Indeed, quiescent MuSC exhibit translational control of visions of the Animals (Scientific Procedures) Act 1986, lineage determinants [27], with Myf5 transcripts held in as approved by the Ethical Review Process Committee of non-translating mRNPs [17, 23]. King’s College London. Prompted by our observation that Fmrp mRNA ex- EDL muscles were dissected from hind limbs of 2–7- pression is induced in G0 [1], in the present study we month-old mice of either sex (two C57BL/6 mice aged profiled expression of a set of mRNP proteins in muscle 8–10 weeks (Fig. 1) or six Pax7-nGFP mice (four males cells and explored their function in quiescence. We first aged 3 months) and two females aged 5 months (Fig. 2) surveyed the expression and distribution of mRNP com- [29]. Isolated muscles were digested with 400 U/ml Type plex proteins in MuSC versus myofibers in isolated sin- I collagenase (Worthington) in DMEM at 37 °C, till sin- gle muscle fiber preparations ex vivo. We report the gle fibers dissociated. All dissociated fibers were trans- enrichment of Fmrp bodies in MuSC in wild-type mice ferred into fresh DMEM medium and triturated gently and reveal a role for Fmrp in MuSC function in vivo to release individual fibers using fire-polished pasteur Roy et al. Skeletal Muscle (2021) 11:18 Page 4 of 28 Fig. 1 (See legend on next page.) Roy et al. Skeletal Muscle (2021) 11:18 Page 5 of 28 (See figure on previous page.) Fig. 1 Expression and distribution of mRNP granule proteins in isolated skeletal muscle fibers.A Schematic of an isolated myofiber (MF) depicting myonuclei (MN) and an associated muscle satellite cell (MuSC). The longitudinal striations represent orientation of the myofibrils while the cross- striations represent the A-band (A) and Z line (Z). B–D, B’–D’ Depict magnified views of the regions enclosed by brackets (dotted lines) in B–D to visualize subcellular distribution of Fmrp. Arrow heads indicate cytoplasm, double arrows indicate nucleus. Fmrp puncta are observed both in nucleus and cytoplasm of MuSC (B, B’) and as cross-striated staining in myofiber. Puncta also accumulate in a cytoplasmic domain adjacent to the MN, while MN is itself not stained (C, C’). Nuclear accumulation of Fmrp is also seen in the Pax7 MuSC nucleus (D, D’) but not in an adjacent Pax7 MN. B’, C’, and D’ represent single-channel (488) images. E–G Distribution of Fmrp puncta (green) in myofiber in a cross-striated pattern congruent with Z lines revealed by α-actinin (red). Arrows in G point to Fmrp puncta co-localizing with α-actinin striations. H–I. Secondary antibody controls (mouse and rabbit) do not show either punctate or striated background. K Distinct GW182 bodies are visible in Pax7 MuSC. Pax7 MN also show distinct perinuclear puncta (arrowheads) and significant punctate staining is observed in MF cytoplasm in a doublet striated pattern likely reflecting A-band localization. K’ Region within brackets in k magnified to show GW182 puncta in the MuSC nucleus (double arrow). L, L’ No enrichment (either nuclear or cytoplasmic) is detected of Dcp1a in MuSC nucleus (marked with the membrane marker Caveolin 1). Faint fibrillar puncta are observed in myofibers. M, M’ Xrn1 is faintly detected in MuSC, but strongly expressed in myofibres in both a longitudinal and cross-striated pattern. K’, L’, and M’ represent single channel (green) images of enlarged areas indicated by brackets in K, L, and M, respectively pipettes. Dispersed single fibers were either immediately mouse per genotype), minced, and digested in collage- fixed in in 4% paraformaldehyde (PFA) for 10 min or nase type II (Cat# LS4196 Worthington Biochemical, cultured for up to 48 h in DMEM, 10% Horse Serum, 400 U/ml final concentration) for 90 min at 37 °C with 20% Fetal Bovine Serum, and 2% Chick Embryo Extract gentle vortexing after every 15 min. The digested muscle (Sera Lab, CE-650-J) followed by fixation, washed three slurry was filtered through 40-μm nylon mesh. The sin- times with PBS, picked and placed on charged slides gle cell suspension was treated with 0.8% ammonium (Thermo-Fisher) for immunostaining. Fibers were chloride to lyse RBCs. Muscle mononuclear cells were permeabilized with 0.5% Tween-20 in PBS for 1 h, washed twice with PBS and stained with biotinylated blocked in 5% BSA in PBS 0.5% Tween-20 for 1 h. Sub- anti-VCAM-1 (BD Biosciences, Cat#553331) primary sequent steps were as for cultured cells. Samples were antibody for 30 min, washed with PBS and stained with imaged on a LSM510 Meta or LSM 880 Airy Scan Streptavidin, Alexa Fluor-488 conjugate (Invitrogen, (Zeiss). Image analysis was done using ImageJ. Cat#S-11223) and CD45-PE (BD Biosciences, Cat#553081) conjugated antibody. Cell sorting was per- Muscle histology formed on Moflo XPD cytometer using gates for the + - To determine muscle cross-sectional area, 2 male adult VCAM-1 and CD45 population. The gated cell popula- -/- mice (6–8 weeks) each for Fmr1 [30] and age-matched tion was sorted directly into growth medium for subse- WT mice were used. The TA muscle was carefully dis- quent culturing on Matrigel (BD Biosciences, sected intact, fixed for 2 h in 4% PFA at 4 °C and equili- Cat#354230) coated dishes for 6 days. brated overnight in 20% sucrose at 4 °C. The muscle tissues were mounted in OCT in cryomoulds and flash Cell culture frozen in a liquid nitrogen-cooled isopentane bath. Serial The C2C12 mouse muscle cell line was cultured in 20 μm cryosections were collected and immunolabelled DMEM supplemented with 20% FBS (with Penicillin/ with anti-laminin antibody to highlight the individual Streptomycin). To generate synchrony in G0 (reversible fiber perimeters, and imaged by confocal microscopy. arrest), cells were cultured in methylcellulose suspension For calculating the cross-sectional area (CSA) of myofi- with 20% FBS for 48 h [32–34]. To differentiate cells into bers, 7 sections were chosen at random from wild-type irreversibly arrested multinucleated myotubes, myoblasts muscle and a corresponding section selected from Fmr1 at 80% confluency were cultured in low serum media -/- muscle (from the equivalent position in the TA). The (DMEM + 2% Horse serum) for 120 h with daily medium CSA of ~ 250 myofibers was measured from confocal changes; myotubes appeared by day 2, with fusion in- images (LSM510 Meta) of sections using ImageJ soft- creasing till day 5. ware, and the mean CSA and two-tailed paired Student’s t test were performed to compare the difference between Western blot analysis the two groups. Cells were lysed in 50 mM Tris HCl, pH 8, 150 mM NaCl, 5 mM MgCl , 0.1% Nonidet P-40) supplemented Isolation of mouse muscle satellite cells with complete protease inhibitor cocktail (Roche Diag- Primary MuSCs were purified from adult mice as de- nostics, France) and incubated on ice for 30 min. Soluble scribed [31]. Briefly, all hind limb muscle groups were proteins were recovered after centrifugation at 15,000 g -/- dissected from 6-week-old WT or Fmr1 mice (1 male at 4 °C for 10 min and quantified by the BCA method. Roy et al. Skeletal Muscle (2021) 11:18 Page 6 of 28 Fig. 2 Dynamics of Fmrp and Dcp1a granules in quiescent and activated MuSCs on isolated single muscle fibers.a–c Immunofluorescence analysis of Dcp1a and Fmrp in freshly isolated EDL myofibers (0 h) and at 12, 24, and 48 h of culture. MuSCs are marked with Pax7-nGFP. a Fmrp puncta are evident in quiescent MuSCs at 0 h of culture, and become dispersed at 24 h, whereas Dcp1a puncta are not evident until 24 h of activation. The patterns of staining are consistent with reciprocal patterns of Dcp1a and Fmrp assembly into granules. b The pattern of Dcp1a in + + + - proliferating MuSCs (Pax7-nGFP EdU ) confirms the timing of MuSC activation. Dcp1a puncta are not observed in quiescent Pax7 EdU cells, but + + + - are present in activated Pax7 EdU MuSC at 24 and 48 h. c By contrast to Dcp1a, the pattern of Fmrp is punctate in quiescent MuSC (Pax7 EdU ) + + and becomes diffuse in activated Pax7 EdU MuSC at 24 and 48 h. Immunofluorescence was performed in N = 3 biological replicates using > 5 EDL fibers for each combination of antibodies in each assay. Scale bars are 10 μm, or 4.8 μm in magnified panels. Proteins were separated on a 4–12% polyacrylamide peroxidase-conjugated goat anti-mouse or anti-rabbit SDS-PAGE gel along with a pre-stained protein ladder IgG—for 45 min at room temperature. After washing in (12–250 kDa) and transferred to a nitrocellulose mem- TBST, the blots were developed using chemilumines- brane. Non-specific protein binding sites were blocked cence solutions and imaged using Image Quant (Anti- by incubation in 5% (w/v) non-fat dry milk (made in body information in Table 3). TBST), for 1 h at room temperature. The membrane was then incubated with antibodies against the different Immunofluorescence and confocal analysis mRNP granule proteins overnight at 4 °C. After washing Cells were grown on cover slips placed in 6-well plates in TBST, the blot was incubated with horseradish (5 × 10 cells per 18-mm cover slip). The next day, cells Roy et al. Skeletal Muscle (2021) 11:18 Page 7 of 28 were rinsed with ice-cold PBS and fixed with 4% PFA GAAGGGACGUUAUUUGUAU. for 10 min at room temperature followed by Non-targeting siRNA pool#1: permeabilization with 0.1% Triton X-100. The cells were UAGCGACUAAACACAUCAA, subjected to immunofluorescence staining with different UAAGGCUAUGAAGAGAUAC, antibodies overnight at 4 °C. The cells were washed with AUGUAUUGGCCUGUAUUAG, cold PBS and incubated with anti-Rabbit Alexa 488 AUGAACGUGAAUUGCUCAA. (Invitrogen A11034, 1:500) and anti-Mouse Alexa 568 Quantitative real-time RT-PCR was performed on an (Invitrogen A11037, 1:500) secondary antibodies at room ABI 7900HT thermal cycler (Applied Biosystems) using temperature for 1 h. The cells were examined by con- the SDS 2.1® ABI Prism software. cDNA was prepared focal microscopy using LSM510 (Zeiss, Germany). Co- from 1 μg total RNA using superscript II (Invitrogen) localization was quantified using ImageJ software, after and used in SYBR-Green assay (Applied Biosystems). maximizing the intensity from all the Z-stacks for a par- Each sample was isolated from three independent bio- ticular image, and analyzed by selecting the puncta as logical samples and analyzed in triplicate reactions. ROI. Image intensity was calculated using Fiji (ImageJ) Amplicons were verified by dissociation curves and se- software and corrected mean intensity (CMI = Total in- quencing. Primer sequences are listed in the Supplemen- tensity of signal − Area of signal × Mean background tary Information. Relative abundance of different signal) was calculated for more than 75 cells. All data mRNAs in Fmrp and Dcp1a knockdown cells was calcu- points were plotted in Box and whisker plot, and p value lated with reference to cells transfected with non- was calculated by two-tailed paired Student’s t test. targeting siRNA and normalized to GAPDH levels. Fold change was calculated using differences in normalized −ΔΔct RNA interference using siRNA cycle threshold value 2 . The following small interfering RNAs (siRNAs) from Primer sequences used in this study: Dharmacon, Thermo Scientific were used for the study: Gapdh - F: 5’-ATCAACCGGGAAGCCCATCAC -3’ siGENOME SMARTpool siRNAs against mouse Fmrp R: 5’- CCTTTTGGCTCCACCCTTCA- 3’ (M-045448-01-0005), mouse Dcp1a (M-065144-01- Cyclin D1- F: 5’-AAGTGCGTGCAGAAGGAGATTG 0005) and non-targeting siRNA pool #1 (D-001206-13- TG-3’ 20); each pool represents 4 distinct siRNAs targeting dif- R:5’ TCGGGCCGGATAGAGTTGTCAGT-3’ ferent sequences in the same transcript. C2C12 myo- Cyclin A2- F: 5’-TTCTGGAAGCTGACCCATTC-3’ blasts maintained in growth medium (DMEM + 20% R: 5’-GGCAAGGCACAATCTCATTT-3’ Fetal bovine serum) were transfected with the siRNAs Cyclin B1- F: 5’-ATGGACACCAACTCTGCAGCAC-3’ listed above using the Lipofectamine ® RNAiMAX Re- R: 5’-CTGTGCCAGCGTGCTGATCT-3’ agent (Invitrogen) according to the manufacturer’s in- Cyclin E1- F: 5’-TGTCCTCGCTGCTTCTGCTTTG structions. Eighteen hours post-transfection, the cells TATCAT-3’ were either induced to differentiate in low mitogen R: 5’-GGCTTTCTTTGCTTG GGCTTTGTCC-3’ medium (DMEM + 2% Horse serum), for 2 days to form Dcp1a -F:5’- CCAGCTGAAGCTCCTACCAC-3’ myotubes (MT) or were synchronized to G0 in suspen- R:5’- CTGTGGGGTCAACCTGAGTT-3’ sion cultures (1.3% methylcellulose in growth medium) Fmr1 - F: 5’-AGGCTTGGCAGGGTATGGTA -3’ [33]. siRNA-transfected cultures were harvested 48 h R:5’-TGTACGATTTGGTGGTGGTCT-3’ after induction of myogenesis or quiescence and sub- Fmr1 -F: 5’-AGAGGAGGAGGCTTCAAAGG-3’ jected to different analyses including EdU proliferation R: 5’- AGAGGAGGAGGCTTCAAAGG-3’ assay, western blotting for different cell cycle proteins Myogenin- F: 5’-TGGGCATGTAAGGTGTGTAAGA-3’ (such as Cyclins A, B, D1, and E), p27, p21, and qRT- R: 5’-ACTTTAGGCAGCCGCTGGT-3’ PCR analysis. Knockdown in these cells was confirmed Pax7 - F: 5’-CATGGTGGGCCATTTCCACT-3’ by western blotting using anti-Fmrp and anti-Dcp1a R: 5’-GGCCCGGGGCAGAACTAC-3’ antibodies (Table 3). p27 - F: 5’-TGCAGTCGCAGAACTTCGAA-3’ siRNA target sequences (smart pool of 4 siRNAs per R: 5’-ACACTCTCACGTTTGACATCTTCCT-3’ transcript): p16 - F: 5’-CGAACTCGAGGAGAGCCATC-3’ Fmrp: GAUUAUCACCUGAACUAUU, R: 5’- CGTGAACGTTGCCCATCATC-3’ GAUCUGAUGGGUUUAGCUA, Myf5 - F: 5’-CCCCACCTCCAACTGCTCTG-3’ CGUCACUGCUAUUGAUUUA, R: 5’-CCAAGCTGGACACGGAGCTT-3’ GAUCAUUCCCGAACAGAUA. MyoD1 - F: 5’-AGCGTCTCGAAGGCCTCAT-3’ Dcp1a: CAACAGCUAUGGGUCUAGA, R: 5’-AGCGCAGCTGAACAAGCTA-3’ GACAGUAGAAGAGUUAUUU, Ki67 - F: 5’-TGGAAGAGCAGGTTAGCACTGT-3’ GUAUAGAAAUGCAAGUUUG, R: 5’-CAAACTTGGGCCTTGGCTGT-3’ Roy et al. Skeletal Muscle (2021) 11:18 Page 8 of 28 EdU incorporation analysis buffer composition: 10 mM Tris-Cl (pH 7.4), 150 mM EdU incorporation was performed in muscle fibers and KCl, 10 mM MgCl , 1 mM DTT, 100 μg/ml Cyclohexi- cell cultures according to the manufacturer’s protocol mide) and samples were centrifuged at 39,000 rpm for (Invitrogen EdU assay Kit Catalog No. C10340). 1.5 h at 4 °C in a SW41 Ti rotor (Beckman Coulter, Brea, CA, USA). The density-separated lysate was analyzed in OPP incorporation assay a polysome profiler linked to a fraction collector (ISCO) OPP incorporation was performed in C2C12 cells cul- with a UA-5 UV detector. Nine fractions of 1 ml were tured for different conditions according to the manufac- collected for each gradient and were used for either pro- turer’s protocol (Invitrogen OPP assay Kit Catalog No. tein analysis by immunoblot or RNA analysis by qRT- C10456). Cells were imaged using Leica SP8 TCS. Inte- PCR. IDAQ software was used for profile generation. grated fluorescence intensities were calculated using Cell The area under each ribosomal peak (40S, 60S, 80S, and Profiler. R was used to perform Multivariate ANOVA polysomes) was calculated using Microsoft Excel and the followed by Tukey’s HSD post hoc test, and ggplot was average area of polysomes was divided by the average implemented for generating box and whisker plots. area of monosomes (80S ribosomes) for each profile in Estimating fluorescence intensities in MT involved ap- order to calculate polysome/monosome (P/M) ratio. proximations in ROI (boundary) selections for the pur- For isolation of RNA from polysome profiles [21, 35], pose of quantification, as MT are multinucleated and we pooled fractions that constituted the mRNPs (frac- overlapping. tions 1 & 2), monosomes (4 & 5), and polysomes (7 & 8) from each gradient derived from CHX- and Puro-treated Apoptosis assay MB, G0, and MT cells, and isolated RNA using TRIzol™ To determine whether Fmrp knockdown cells undergo LS Reagent (Invitrogen) according to the manufacturer’s apoptosis, we used the Invitrogen Apoptosis Kit as per instructions. The isolated RNA was resuspended in the manufacturer’s protocol (Cat no. V13245). Flow cy- equal volumes of water, quantified and checked for its tometry was performed on a BD Fortessa, using FlowJo purity by Nanodrop. Equal volumes of RNA from the software for analysis. pooled fractions were subjected to cDNA synthesis using SuperScript IV (Invitrogen) according to the manufac- Senescence assays turer’s protocol. Quantitative real-time PCR analysis was Control or Fmrp knockdown cells were evaluated for ex- performed using Power SYBR Green (Applied Biosys- pression of p16 and p21 by qRT-PCR. Activity of tems) in ABI 7900HT Thermal cycler (Applied Biosys- senescence-associated β-galactosidase (β-Gal) enzyme tems). Serial dilutions of cDNA resulting in decrease in was tested through X-gal staining assay at pH 6.0 to sup- copy number were used to generate a standard curve for press lysosomal β-gal activity. Briefly, cells were fixed each gene. C values of each dilution were plotted with 4% PFA for 10 min at room temperature, washed against the copy numbers (ln of dilution). For experi- twice with PBS, and incubated with the chromogenic mental samples, the copy number was calculated using substrate 5-bromo-4-chloro-3-indolyl-beta-d-galactopyr- C value and standard curve obtained for that mRNA anoside (X-Gal) staining solution comprising 1 mg/ml with the same set of primers. The percentage of total X-gal (in DMF), 5 mM potassium ferrocyanide, 5 mM mRNA across the gradient was calculated as follows: potassium ferricyanide and 1 mM MgCl , overnight at (copy number in specific fraction/Total copy number of 37 °C. DNA damage-induced foci of γH2AX were de- the gene across the gradient) × 100. tected by immunofluorescence as detailed in the Supple- mentary Information. Results mRNP components are differentially expressed in Polysome analysis quiescent muscle stem cells versus myofibers Fifteen million cells (MB, MT, or G0) were incubated To investigate mRNP protein distribution in muscle, we with 0.1 mg/ml of Cycloheximide for 15 min or with 0.1 used isolated murine myofibers ex vivo, complete with mg/mL of Puromycin for 2 h prior to lysis. Cells were resident MuSCs in their niche. At a subcellular level, lysed in ice-cold lysis buffer (10 mM Tris-Cl (pH 7.4), mRNP granule components are known to partition be- 150 mM KCl, 10 mM MgCl , 1 mM DTT, 100 μg/ml Cy- tween a diffuse cytoplasmic distribution and punctate TM cloheximide, 1% NP40, 1× cOmplete EDTA-free Pro- granules, where puncta represent functional complexes tease inhibitor (Cat. No. 5056489001), RNAse inhibitor of RNA and proteins [36, 37]. Immunofluorescence ana- (Cat. No. 10777-019), 6 U/ml). After incubation in cold lysis revealed that mRNP granule proteins are organized lysis buffer for 30–45 min, the lysate was spun at in puncta that were enriched in quiescent Pax7 MuSC 13000×g for 15 min at 4 °C. The supernatant was then (Fig. 1). Myofibers also showed puncta, organized in stri- loaded on to 10–45% (wt/wt) sucrose gradient (gradient ated patterns that suggest association with underlying Roy et al. Skeletal Muscle (2021) 11:18 Page 9 of 28 -/- cytoskeletal elements. Specifically, the translational re- knockout (Fmr1 ) mouse. Quantification of cross- pressor Fmrp showed punctate immunolabeling that was sectional area of muscle fibers in cryo-sections of adult highly enriched in the cytoplasm of Pax7 MuSC (Fig. tibialis anterior muscle revealed that muscle fibers in -/- 1B–D), but also associated with sarcomeres in myofibers Fmr1 muscle showed drastically reduced caliber [mean (Fig. 1E–G), with distinct non-sarcomeric enrichment in ± SD of 619 μm ± 200] compared with age-matched - 2 the cytoplasm adjacent to Pax7 myonuclei (Fig. 1C, C’). wild-type (WT) mice [mean ± SD of 1518 μm ± 438, n Interestingly, Fmrp was also located in MuSC nuclei = 250 fibers; p value < 0.0001] (Fig. 3a, b). As there is (Fig. 1B, B’,D,D’), but not in myonuclei. Another trans- some expression of Fmrp in myofibers, a direct effect of lational repressor GW182, which is involved in the Ago- this mRNP granule protein in a fiber-intrinsic mechan- miRNA pathway, showed a similar distribution to Fmrp: ism cannot be ruled out. However, we also found that + -/- - + discrete cytoplasmic puncta in Pax7 MuSC and in FACS-isolated Fmr1 CD45 VCAM-1 MuSCs (Fig. 3c) zones near myonuclei, with smaller puncta in myofibers, proliferate less compared to WT controls (Fig. 3d). The -/- arranged in a distinct pattern reflecting sarcomeric proportion of sorted MuSCs was similar in Fmr1 and organization (Fig. 1 K). Thus, proteins implicated in WT (Fig. 3c), by contrast to an earlier report by Fujita +- translational repression are located in mRNP granules et al. [23], who reported that the number of Pax7 - -/- clearly evident in quiescent MuSC. MyoD cells on single fibers were lower in Fmr1 mice. To determine the distribution of proteins involved in This difference may reflect differences in the methods mRNA turnover, we examined expression of key regula- and markers used to identify MuSCs, as well as the tors of mRNA, the decapping enhancer Dcp1a and the numbers of cells analyzed in the two studies. Further, we 5’-3’ exoribonuclease Xrn1 [14]. Dcp1a protein was not found that when equal numbers of sorted WT and -/- -/- detected in quiescent MuSC (marked by MuSC-enriched Fmr1 MuSCs were plated, there were fewer Fmr1 membrane protein Caveolin 1 (Cav) (Fig. 1L), but cells over the course of 6 days in culture compared to formed a fine striated pattern in the myofiber cytoplasm, WT (Fig. 3d). Consistent with this observation, acute largely perpendicular to expected sarcomeric knockdown of Fmrp in C2C12 myoblasts reduced clono- organization. Similarly, Xrn1 was not present in MuSC, genic performance (Fig. S3A) but did not lead to in- but exhibited a clear striated pattern in myofibers (Fig. creased cell death through apoptosis (Fig. S3B), nor were 1M). These observations reveal that components of the markers of senescence significantly induced (Fig S4). To- mRNA storage/stabilization complex (Fmrp, GW182) gether, these preliminary results indicate that Fmrp ex- are highly expressed in the nuclei of quiescent MuSC, pression is required for achieving normal fiber caliber in while the mRNA decay complex components (Dcp1a, postnatal adult skeletal muscle and that this phenotype Xrn1) are not. may be linked to a defect of knockout MuSC in prolifer- ation, reactivation from quiescence or clonogenic self- Reciprocal dynamics of Fmrp and Dcp1a puncta during renewal in culture. MuSC activation on single fibers To examine the dynamics of mRNP granules during Differential expression of mRNP proteins in quiescent, MuSC activation, we cultured isolated myofibers for up proliferating, and differentiated muscle cells in culture to 48 h and determined the pattern of expression of ,To explore the muscle cell-intrinsic functions of Fmrp, Fmrp and Dcp1a. Activation of MuSCs led to EdU in- we examined the expression of a series of mRNP pro- corporation at 24 and 48 h. Fmrp puncta were promin- teins (schematized in Fig. 4a), in a tractable adult ent in quiescent MuSC, but showed dispersal and a MuSC-derived murine C2C12 culture model that per- diffuse staining pattern after 24 h. By contrast, Dcp1a mits the generation of pure populations of proliferating puncta were not seen in quiescent MuSC, and became myoblasts (MB), quiescent myoblasts (G0), or differenti- evident in activated and proliferating MuSC (EdU+) at ated myotubes (MT) [2, 26]. In particular, this model al- 24 h (Fig. 2a–c). Thus, activation of MuSCs in their lows the entry to mitotic quiescence to be examined, a myofiber niche is associated with reciprocal changes in limitation in other similar techniques. The three cellular accumulation of mRNP granules associated with oppos- states were distinguished using expression of Myogenin ing functions, i.e., transcript turnover (Dcp1a) vs. (Myog), a master regulator of myogenic differentiation, utilization (Fmrp). and Cyclin D1, a canonical marker of proliferation: MB + - - + are Cyclin D1 Myog , MT are Cyclin D1 Myog and - - Fmrp knockout mice exhibit altered muscle stem cell G0 are Cyclin D1 Myog (Figs 4b and S5A). We investi- proliferation gated the abundance of Fmrp and Dcp1a, together with To examine whether Fmrp observed in mRNP granules other categories of mRNP components, namely: (i) pro- in quiescent MuSC is important for stem cell function teins involved in translation repression and formation of in vivo, we analyzed skeletal muscle from the Fmr1 SGs (Fmrp, Tia1) [10, 40], and translation initiation Roy et al. Skeletal Muscle (2021) 11:18 Page 10 of 28 -/- Fig. 3 ,Reduced muscle fiber caliber and MuSC proliferation in Fmr1 mice. a Cryo-sections (20 μm) of adult tibialis anterior muscle isolated from -/- wild-type (WT, left) and Fmr1 mice (right) immunolabelled with laminin (red) and nuclei counterstained with DAPI (blue): myofibers show -/- reduced diameter in Fmr1 muscle. b Left panel shows qRT-PCR quantification of mRNA encoding Fmrp isolated from whole muscle of adult WT -/- and Fmr1 mice. Values represent mean + SEM; n = 2, two-tailed paired Student’s t test is indicated. ** p < 0.01. Fmr1 RNA is detectable at lower -/- -/- levels in Fmr1 as described [30, 38, 39]. Right panel: quantification of mean myofiber cross-sectional area (CSA) in wild-type and Fmr1 muscle cryosections. Values represent mean + SD; n = 250. CSA from two mice. Two-tailed unpaired Student’s t test is indicated. *** p value < 0.0001, (N -/- + = 2 male mice per genotype). c–d. Muscle stem cells isolated from Fmr1 mice do not proliferate well in culture. c The proportion of VCAM , - -/- - + CD45 MuSC is similar in adult WT and Fmr1 mice. However, there is a noticeable reduction in the VCAM , CD45 cells suggesting effects on the -/- leukocyte compartment. d Equal numbers of FACS purified MuSC isolated from the hind limb muscle of adult WT and Fmr1 mice were plated -/- in culture for 0 or 6 days. Fmr1 cells show poor population expansion (N = 1 male mouse per genotype) (eIF-4E), (ii) proteins involved in the PB nonsense- but in MT, Fmrp was downregulated. Consistent with mediated mRNA decay pathway (Dcp1a, Pat1, and reversible suppression of translation in G0, eIF-4E, the Edc4), and (iii) proteins known to shuttle between these cap-binding component of the rate limiting translation two complexes, (Xrn1, Gw182, and Ago2) [10, 41] (Fig. initiator eIF-4F complex, is strongly downregulated in 4c, d). Briefly, the translation repressors Fmrp and Tia1 G0, but upregulated in MT. Dcp1a (and another mRNA continued to be expressed in G0 at levels similar to MB, decay factor Edc4) were less abundant in both G0 and Roy et al. Skeletal Muscle (2021) 11:18 Page 11 of 28 Fig. 4 Differential expression of mRNP granule proteins in proliferating, quiescent and differentiated muscle cells in culture. a Schematic depicts segregation of transcripts into translating and non-translating pools on emergence from the nucleus with a constellation of RNA-binding proteins. Non-translated transcripts may be sequestered in mRNPs enriched for decay complex (mRNA turnover) or storage granule components (translational repression/stabilization of mRNA). b Western blot analysis showing that three distinct cellular states can be distinguished by expression of Myogenin and Cyclin D1 (MB: asynchronously proliferating myoblasts are CycD1+, MyoG-; MT: 5-day differentiated myotubes are CycD1-, MyoG+; G0: quiescent myoblasts are CycD1-, MyoG-). c Western blot profile of mRNP granule protein expression across three cellular states. MB: proliferating myoblasts; G0: quiescent myoblasts; MT: differentiated myotubes. Expression of most proteins is suppressed in G0; notable exceptions are Fmrp and Tia1, which are maintained in G0 (see Table 1 and quantification in Fig. 4d). d Quantification of relative expression of mRNP granule proteins across the three cellular states calculated from densitometric analysis of immuno-blots. Each protein was normalized to Gapdh in the same sample, before normalizing to MB. Values represent mean + SD, n = 3, two-tailed paired Student’s t tests are indicated as [* p ≤ 0.05, ** p ≤ 0.01] Roy et al. Skeletal Muscle (2021) 11:18 Page 12 of 28 MT compared with MB. Overall, the quantitative ana- puncta re-appeared, consistent with the abundance of lysis of mRNP granule protein expression (Fig. 4c, d) re- Dcp1a protein in cycling MB (Fig. 5a). In MT, Fmrp was vealed that when comparing G0 to MB, proteins organized as small, dispersed cytoplasmic granules, while involved in mRNA turnover such as Dcp1a and Edc4, Dcp1a puncta were reduced compared to MB (Fig. 5a). were under-represented, while proteins involved in There was a general similarity in abundance of puncta in mRNA storage/stabilization/translational stalling (Fmrp, culture and in vivo: i.e., similar patterns in MT and myo- Tia1) were maintained at similar levels in G0 when com- fiber versus G0 and MuSC, suggesting an association of pared to MB. particular mRNP granule dynamics with these distinct To assess whether changes in expression of mRNP cellular states. The degree to which three additional proteins resulted from changes in expression of their mRNP proteins (Edc4, Pat1, Ago2) were organized into mRNAs, we used bioinformatic analysis of available puncta also varied between cellular states (Fig. S1). Add- transcriptome data from in vivo fixed (quiescent) satel- itional representative images (Fig. S2A) show the in- lite cells compared with activated (proliferating) satellite creased presence of Fmrp puncta in G0 and Dcp1a cells [42] and grouped them according to their function puncta in proliferation, which is supported by quantifi- [43, 44]. This comparison revealed that expression of cation of mean fluorescence intensities of these proteins genes encoding translational stalling complex proteins in subcellular puncta (Fig. S2B), and confirms the recip- (Tia1, Tncr6b, Ddx5, Ddx17) are upregulated in G0, rocal expression/organization of Fmrp and Dcp1a in G0 whereas genes encoding proteins important for mRNA compared to proliferating cells. Taken together, these decapping and turnover (Dcp1a, Edc4, Edc3) are down- immuno-localization studies indicate that translational regulated in G0 (Table 1). Fmrp expression was found to repression complexes (Fmrp, Ago2) are more prominent be maintained in quiescent satellite cells. Taken to- in G0 than in MB and MT, and that nonsense-mediated gether, this analysis suggests that compared with prolif- mRNA decay complexes (Dcp1a, Edc4, Pat1) are more erating MB, the decapping machinery is suppressed in prominent in MB and MT than in G0, both of which are G0, but that translational repression capacity is main- consistent with earlier reports of transcript stabilization tained/enhanced in G0. in quiescent cells [32, 34, 46]. Distinct organization and dynamics of mRNP granules in Global translation rates and expression of translation two mitotically inactive states initiation factors are suppressed in G0 To compare the distribution and dynamics of mRNP To compare global translation rates between the prolif- complexes in different cellular states in culture, we ex- erating, differentiated and quiescent states, we used in- amined the staining pattern of Fmrp and Dcp1a using corporation of O-propargyl-puromycin (OPP) to immunofluorescence confocal microscopy (Fig. 5a). As biosynthetically label nascent proteins (Fig. 5b, c). Rap- active mRNPs self-assemble into observable puncta and idly growing MB pulsed with OPP for 1.5 h showed disassemble upon releasing bound mRNA [36, 43], sub- strongly labeled cytoplasm and nucleoli, possibly reflect- cellular staining patterns are a reflection of the activity ing the nucleolar location of newly synthesized riboso- state of these complexes. In asynchronously proliferating mal proteins during ribosome biogenesis. In fused MT, MB, Dcp1a and Fmrp were present in small, numerous, cytoplasmic OPP labeling predominated, likely reflecting non-overlapping cytoplasmic puncta, consistent with greater synthesis of sarcomeric and other non-ribosomal their participation in distinct complexes with distinct proteins. By contrast, G0 cells showed low and variable functions. In G0, whereas Dcp1a immunolabeling was OPP labeling of the cytoplasm. Many G0 cells were es- low and diffuse (not punctate), the size and intensity of sentially unlabeled above background levels (Fig. 5b, c). cytoplasmic Fmrp granules dramatically increased, and Moreover, nucleoli could not be distinguished (Fig. 5b). nuclear-localized Fmrp granules were also prominent, These findings indicate lower rates of protein synthesis while total Fmrp protein level was maintained (Fig. 4c, and ribosome assembly in G0 cells. d), suggesting enhanced granule assembly, and greater To investigate translation by an independent method, translational repression. Notably, mRNP immuno- we analyzed expression of two translation initiation fac- detection patterns in cultured G0 cells (Fig. 5a) reflected tors, eIF-4E (the rate limiting factor in cap-dependent the patterns observed in vivo in MuSC (Fig. 1) with re- translation that also regulates mRNA export) and eIF4G spect to (i) increased Fmrp puncta and reduced Dcp1a (a scaffold for assembly of the eIF-4F complex compris- puncta and (ii) the appearance of Fmrp puncta in the G0 ing eIF-4E, eIF-4G, and eIF-4A on the 5’ cap). Both pro- nucleus. teins were present in MB and MT, but both were much ,We next tested the effect of cell-cycle reactivation on reduced in G0 (Fig. 5d), and strongly re-induced in a mRNP granules. Three hours after reactivation from qui- punctate pattern after reactivation from quiescence (R3), escence (R3) Fmrp puncta disappeared and Dcp1a when protein synthesis begins to recover (Fig. 5 c, d). Roy et al. Skeletal Muscle (2021) 11:18 Page 13 of 28 Table 1 Bio-informatic analysis of transcripts encoding mRNP components Data from Yue et al (2020) Accession GSE113631 mRNP component genes FPKM values fSC vs ASC Table 1 Functional Class Gene ASC_ ASC_ fSC_ fSC_ Fold change adjusted p-value (BH Fold Symbol rep1 rep2 rep1 rep2 (Log2) correction) change ARE binding Tnpo1 57.211 65.710 17.063 15.135 -1.93 0.0353 0.26 ARE binding Zfp36 22.033 19.558 15.694 13.194 -0.53 0.1265 0.69 ARE binding Tia1 7.034 6.693 16.816 16.464 1.28 0.0117 2.42 Deadenylation Cnot1 105.514 110.661 20.399 21.317 -2.37 0.0135 0.19 Deadenylation Cnot2 41.603 42.019 26.262 15.837 -0.99 0.1112 0.50 Deadenylation Cnot6l 32.610 34.390 24.860 25.916 -0.4 0.0481 0.76 Deadenylation Cnot8 25.708 26.029 30.524 23.093 0.05 0.8439 1.04 Deadenylation Pan2 12.528 10.940 15.276 14.178 0.33 0.1543 1.26 Deadenylation Tob2 11.102 12.033 16.704 14.161 0.42 0.1717 1.34 Deadenylation Cnot4 10.540 8.787 11.628 17.230 0.58 0.3379 1.49 Deadenylation Pan3 3.337 3.622 5.453 5.373 0.64 0.0281 1.56 Deadenylation Cnot3 0.864 0.889 3.501 3.168 1.93 0.0252 3.81 Decapping Lsm3 249.390 284.670 85.416 78.982 -1.7 0.0357 0.31 Decapping Dcp1a 16.651 17.486 5.447 5.466 -1.65 0.0151 0.32 Decapping Edc3 12.841 12.809 5.918 4.803 -1.26 0.0274 0.42 Decapping Patl1 10.775 9.751 5.821 4.656 -0.97 0.0601 0.51 Decapping Edc4 24.210 23.716 12.896 12.523 -0.91 0.0130 0.53 Decapping Dcp2 9.174 10.101 7.586 8.180 -0.29 0.1489 0.82 Decapping Lsm2 32.104 34.063 28.098 31.387 -0.15 0.3121 0.90 Decapping Lsm6 209.767 214.354 204.279 179.065 -0.15 0.3451 0.90 Decapping Lsm7 58.943 65.467 66.429 58.668 0.01 0.9582 1.01 Decapping Lsm5 87.274 87.571 126.064 111.808 0.44 0.0971 1.36 Decapping Lsm1 4.656 4.518 7.664 5.746 0.55 0.2374 1.46 Decapping Dcp1b 0.440 0.394 2.564 2.611 2.63 0.0081 6.20 Decapping Lsm4 5.825 5.767 42.218 38.352 2.80 0.0207 6.95 Many RNP functions Eif4e 230.579 234.878 57.512 38.165 -2.28 0.0202 0.21 Many RNP functions Ddx1 56.948 56.634 16.101 9.526 -2.15 0.0276 0.23 Many RNP functions Eif2s2 183.202 191.179 68.677 56.697 -1.58 0.0214 0.33 Many RNP functions Eif2b1 76.122 77.151 33.579 23.721 -1.42 0.0379 0.37 Many RNP functions Gemin5 70.000 66.713 28.690 30.406 -1.21 0.0184 0.43 Many RNP functions Trim59 4.600 4.463 2.249 1.956 -1.11 0.0247 0.46 Many RNP functions Lima1 15.817 12.996 8.282 10.047 -0.65 0.1513 0.64 Many RNP functions Dhx40 5.881 6.893 5.668 4.463 -0.33 0.3256 0.79 Many RNP functions Pdlim7 15.903 17.440 19.606 18.779 0.20 0.1690 1.15 Many RNP functions Fmr1 42.281 42.132 58.735 49.227 0.35 0.2056 1.28 Many RNP functions Xrn1 11.279 11.434 14.211 17.589 0.49 0.1853 1.40 Many RNP functions Eif2b4 22.085 21.683 33.687 31.827 0.58 0.0322 1.50 Many RNP functions Pcbp2 77.519 73.659 138.630 159.083 0.98 0.0542 1.97 Many RNP functions Trim28 19.447 19.682 33.020 45.418 1.00 0.1500 2.00 Many RNP functions Lpp 30.789 32.101 113.147 97.692 1.75 0.0385 3.35 Many RNP functions Ddx5 65.868 58.928 263.633 201.650 1.90 0.0746 3.73 Roy et al. Skeletal Muscle (2021) 11:18 Page 14 of 28 Table 1 Bio-informatic analysis of transcripts encoding mRNP components (Continued) Data from Yue et al (2020) Accession GSE113631 mRNP component genes FPKM values fSC vs ASC Table 1 Functional Class Gene ASC_ ASC_ fSC_ fSC_ Fold change adjusted p-value (BH Fold Symbol rep1 rep2 rep1 rep2 (Log2) correction) change Many RNP functions Peg3 105.426 104.905 445.266 466.449 2.12 0.0135 4.33 Many RNP functions Ybx1 64.225 64.993 464.325 296.992 2.56 0.1191 5.90 Many RNP functions Ddx17 9.400 8.959 111.803 114.915 3.63 0.0081 12.35 Many RNP functions Trim21 0.015 0.017 0.417 0.613 5 0.0809 32.00 miRNA-mediated gene Limd1 46.125 42.266 43.565 43.929 -0.01 0.8565 0.99 silencing miRNA-mediated gene Tnrc6a 8.243 8.496 10.950 13.634 0.55 0.1674 1.47 silencing miRNA-mediated gene Htt 2.976 2.676 6.126 6.307 1.14 0.0195 2.20 silencing miRNA-mediated gene Tnrc6b 4.196 4.958 14.202 15.669 1.71 0.0293 3.27 silencing miRNA-mediated gene Ipo8 4.524 4.272 16.861 15.367 1.87 0.0242 3.66 silencing miRNA-mediated gene Tnrc6c 3.082 3.039 16.137 17.995 2.48 0.0247 5.58 silencing NMD pathway Smg7 58.640 66.881 11.867 12.560 -2.36 0.0299 0.19 NMD pathway Upf2 13.025 15.822 16.257 16.691 0.19 0.3801 1.14 NMD pathway Upf1 14.745 16.093 17.029 20.119 0.27 0.2885 1.21 NMD pathway Pnrc2 44.583 42.606 61.898 49.546 0.35 0.2770 1.28 NMD pathway Upf3b 24.102 22.611 42.912 44.988 0.91 0.0230 1.88 NMD pathway Smg5 2.169 2.287 7.059 6.539 1.61 0.0217 3.05 NMD pathway Smg6 6.228 6.373 26.228 27.657 2.10 0.0148 4.28 NMD pathway Upf3a 5.788 5.261 34.781 29.334 2.54 0.0379 5.82 NMD pathway Pnrc1 3.468 2.819 103.360 88.526 4.93 0.0293 30.52 The altered abundance of eIF-4E was consistent with the maintenance of quiescence [48], the data above sug- our western analysis (Fig. 4 c, d). These initiation fac- gest that a number of G0-induced transcripts may also tors showed some nuclear localization in G0, which be sequestered in non-polysomal compartments, to be was greatly enhanced during synchronous reactivation, mobilized for protein synthesis required for the return but not detected in either cycling MB or MT, pos- to the cell cycle [17]. To visualize ongoing translation sibly reflecting involvement in upstream functions activity directly, we analyzed steady-state polysome pro- such as mRNA export, that are important for cell files in each cellular state. To ensure polysome integrity cycle re-entry (Fig. 5d). Taken together, these findings during isolation and display, cells were treated briefly are consistent with the notion of G0 as a state where with cycloheximide (CHX) prior to lysis, to arrest trans- global translational repression is coupled to mRNA lating ribosomes on mRNAs, followed by separation on stabilization in granules, keeping cells primed for cell sucrose density gradients (Fig. 6a–c). The profile of cycle re-entry [17, 34]. RNA-protein complexes was quantified in density- separated fractions and analyzed by immuno-blotting. A Quiescent myoblasts exhibit puromycin-resistant mRNP second profile was run from cells in each state that were complexes in G0 treated briefly with puromycin (Puro), that successfully ,Quiescent cells show low transcriptional activity com- disengaged mRNA from translating ribosomes, removing pared with proliferative or differentiated states. Although the polysome profile (Fig. 6a–c). With respect to mRNP many transcripts are specifically induced in G0 [1, 47] granule dynamics, CHX rapidly dissociates mRNP gran- and some must be translated into proteins required for ules, whereas Puro promotes their assembly [10]. Roy et al. Skeletal Muscle (2021) 11:18 Page 15 of 28 Fig. 5 Assembly of mRNP into puncta in different cellular states correlates with levels of protein synthesis. a Representative immunofluorescence images showing Fmrp (green) and Dcp1a (red) puncta in G0, MB, and MT, as well as cells reactivated for 3 h from G0 (R3). Arrows indicate prominent puncta. Notably, Fmrp puncta are large and prominent in G0, disperse at 3 h post reactivation and are less evident in asynchronous MB. Dcp1a puncta are nearly absent in G0 and reappear at R3; Dcp1 puncta are also more prominent in MB than MT. Fmrp and Dcp1a mostly localize to distinct puncta; the rare yellow puncta seen in R3 and MT may reflect transient co-localization due to passage of transcripts between two types of mRNP granules as inferred in [45]. b Measurement of the rate of protein synthesis using OPP incorporation into newly synthesized proteins reveals active translation in MB and MT, and substantial suppression in G0. c Quantification of images in b. Fluorescent intensity was measured in 150 cells from each condition, the box and whisker plot shows integrated fluorescence for each cell (each dot represents one cell), limits on the box correspond to 75th and 25th percentile values. “Mb-neg” and “G0-neg” represent samples of MB and G0 that were not pulsed with OPP but processed for detection along with samples that were exposed to OPP. N = 2 biological replicates. Data were analyzed by multivariate ANOVA with post hoc HSD Tukey tests performed for each pairwise comparison. *** p < 0.0001, ** p < 0.001. d Immunolabeling of translation initiation factors eIF-4E (red) and eIF-4G (pink) in G0, R3, MB, and MT: Upper panel shows merged images, and lower panels show detection of each factor individually. Expression and assembly of these translation factors correlates with levels of protein synthesis seen in b and c: poor in G0, restored assembly with distinct puncta in R3, and strong expression and organization of eIF-4E and eIF-4G complexes in MB and MT Roy et al. Skeletal Muscle (2021) 11:18 Page 16 of 28 Fig. 6 Polysome profiles of proliferating, quiescent, and differentiated muscle cells reveal stalled polysomes in G0. Translational profiles of myoblasts (a), myotubes (b), and G0 cells (c) using polysome display on sucrose gradients. Panels on left depict profiles derived from cells briefly treated with CHX to ‘fix’ ribosomes in the act of translation, while panels on the right depict profiles derived from cells treated with Puro to disrupt translation by mRNA release. Western blotting of proteins isolated from 9 individual 1-ml fractions from the sucrose gradients (equal volumes loaded) reveals (i) distribution of ribosomes in each fraction based on ribosomal protein P0 (middle) and the extent of association of decay complex based on Xrn1 (top), and translation inhibitory complex based on Fmrp (botttom) with each fraction. Comparison of the profiles and distribution of individual proteins reveals very poor translation in G0, correlating with the OPP incorporation in Fig. 5. The presence of puromycin-insensitive complexes in G0 arrested cells, suggests polysome stalling. d Analysis of transcript distribution in polysome profiles correlates with rate of protein synthesis and suggests low mRNA utilization in G0. qRT-PCR analysis of selected transcripts (GAPDH, Cyclin D1, MyoD, Myf5, and p27) from RNA isolated from the mRNP-, monosome-, and polysome-containing fractions of profiles depicted in Fig. 6a–c. All transcripts tested show substantial enrichment in the mRNP and monosome compartment in G0 compared with the monosome and polysome fraction, suggesting a severe suppression of protein synthesis consistent with the OPP incorporation study (Fig. 5b, c). Notably, none of the transcripts tested show appreciable enrichment in the mRNP fraction in MB and MT, indicating their robust translational utilization in the polysomal compartment. Values represent the mean + SD of transcript levels in fractions from two independent polysome profiles for each condition Roy et al. Skeletal Muscle (2021) 11:18 Page 17 of 28 In MB, in addition to the strong ribosomal subunit Table 2 Non-translating to translating ratio (mRNP: mono + poly) ratio in 3 different states (MB, MT, and G0) peaks in fractions 3, 4, 5 (containing free 40S, 60S sub- units, and 80S monosomes, respectively), a range of State Transcript polysome peaks was visible (fractions 6–9), which was Cyclin D1 p27 MyoD Myf5 Gapdh sensitive to Puro treatment, showing that the cells were MB 0:100 0:100 0:100 2:98 3:97 engaged in active translation (Fig. 6a). Western blot pro- MT 0:100 0:100 0:100 0:100 0:100 files confirmed that the ribosome-containing heavier G0 70:30 60:40 50:50 65:35 50:50 fractions (3–9, marked by the presence of the ribosomal protein P0) were largely devoid of mRNP proteins Xrn1 and Fmrp, which were enriched in the non-polysomal Cdkn1b/p27), as seen in global transcript analysis de- fractions 1–2. On treatment with Puro, disruption of rived from transcriptome data (Venugopal et al, 2020 polysomes was evident and accompanied by loss of P0 [49]) (Fig. S5B and Table 2). Consistent with the bulk protein from fractions 6–9 (Fig. 6a). MT also displayed polysome profile that shows low polysome assembly in very active translation, showing monosome and poly- G0, all transcripts tested show substantial enrichment in some peaks similar to the profile in MB (Fig. 6b). Simi- mRNP and monosome compartments and < 10% in the larly, Xrn1 and Fmrp were detected at low levels in polysome fraction in G0 cells (Fig. 6d). By contrast, in fraction 6, but otherwise, these mRNP proteins were both MB and MT, all five mRNAs were enriched on largely absent from the polysome fractions 7–9. As in polysomes, with barely detectable presence in the mRNP proliferating MB, Puro treatment led to the loss of poly- fraction, with Fig S5B reflecting the global transcript sta- some peaks in MT. tus of the mRNAs relative to MB consistent with the In G0 cells, by contrast, polysomes were nearly un- high rates of protein synthesis typical of these states. detectable and fewer monosomes were seen, consistent Thus, the observation that, for all five transcripts tested, with the accumulation of P0 in the lighter complexes the majority (as a proportion of each transcript’s abun- (Fractions 1–4 (Fig. 6c). Nevertheless, in G0 cells, P0 dance in each state), was either found in large polysomes persisted in high molecular weight complexes (Fractions in MB (i.e., was actively translated), shifted to mono- 6–9 in Puro vs CHX), that were insensitive to Puro (Fig. somes compared with polysomes in MT (i.e., was less 6c). Together, these observations suggest the presence of actively translated), or further shifted to mRNPs in cells heavy mRNP complexes in G0 cells that are not engaged in G0 (i.e., was not translated at all), we interpret to re- in active translation. These heavy mRNPs could be flect the global translational status of each state. Taken stalled polysomes, or mRNA captured in other heteroge- together with the repressed global rates of protein syn- neous paused complexes along with ribosomes, but not thesis and increased accumulation of mRNP proteins in undergoing active translation. Treatment of G0 cells visible puncta, we conclude that mRNA sequestration in with Puro increased mRNPs in the heavy fractions 7–9, a non-translated compartment is a broad regulatory the opposite of the effect of Puro in MB (Fig. 6a, c). The process that is enhanced in reversible G0, but not in sustained enrichment in the heavier fractions (7–9) in post-mitotic MT. puromycin-treated G0 cells suggests that ribosomal pro- teins are present in non-canonical high molecular weight Dcp1a and Fmrp reciprocally regulate their protein complexes in G0 cells, which are absent in MB and MT. abundance and granule assembly Taken together with OPP incorporation and eIF-4E ex- As Fmrp and Dcp1a are known to regulate distinct as- pression levels, these results demonstrate that proliferat- pects of mRNA function (translation vs. turnover [15]) ing and differentiated cells are actively engaged in and were found in different complexes, we considered translation, while quiescent cells show markedly sup- the possibility that these proteins might also cross- pressed protein synthesis, potentially associated with se- regulate. We used siRNA-mediated knockdown to per- questered and stalled ribosomes. turb the levels of each protein and evaluated the effect of knockdown of one protein on abundance of the other Transcripts accumulate in a non-polysomal mRNP protein using western blotting (Fig. 7a). Proliferating compartment specifically in G0 myoblasts were transfected with siRNA smart pools To probe the distribution of specific transcripts between (comprising four independent siRNAs) designed to tar- actively translating and inactive sequestered compart- get either Dcp1a or Fmr1 mRNAs. A non-targeting ments, we used qRT-PCR analysis on RNA isolated from siRNA pool was used as a control. Knockdown efficiency the mRNP, monosome- and polysome-containing frac- was confirmed to be 70–85% for Fmrp and 40–50% for tions (Fig. 6d). We selected mRNAs whose levels are (i) Dcp1a by western blotting (Fig. 7a). Indeed, knockdown unchanged (Gapdh) (ii) suppressed in G0 (Cyclin D1, of Fmrp led to an induction of Dcp1a protein levels and MyoD), or (iii) maintained/induced in G0 (Myf5, vice versa, knockdown of Dcp1a was accompanied by Roy et al. Skeletal Muscle (2021) 11:18 Page 18 of 28 Fig. 7 (See legend on next page.) higher levels of Fmrp (Fig. 7a). This reciprocal regulation knockdown of Fmrp was readily observed as reduced im- at the level of protein abundance was accompanied by munofluorescence, and accompanied by an enhanced in- increased detection of the respective protein in cytoplas- tensity of Dcp1a, and reciprocally, knockdown of Dcp1a, mic puncta (Fig. 7b). Quantification of the fluorescent led to loss of Dcp1a detection and enhanced intensity of intensity of cytoplasmic staining (Fig. 7c) revealed that Fmrp. Taken together, these experiments reveal cross- Roy et al. Skeletal Muscle (2021) 11:18 Page 19 of 28 (See figure on previous page.) Fig. 7 Cross-regulation of Fmrp and Dcp1a in knockdown myoblasts. a Following transfection with either Dcp1a or Fmr1 siRNA pools or the control pool (Scr), knockdown myoblasts were incubated in growth medium for 18 h. Left: Immunoblot analysis shows that in Fmrp knockdown, Fmrp abundance is reduced but Dcp1a expression is enhanced. Likewise, Fmrp protein levels are increased in Dcp1a knockdown. Values depicted under each lane represent protein levels from normalized densitometric scans, relative to level in Scr. Right panel: Densitometry of western blots of Dcp1a, and Fmrp proteins normalized relative to Scr with Gapdh as internal control. Bar graphs represent mean ± SD from n = 3. Two-tailed paired Student’s t tests are indicated as * p <0.05. ** p <0.01. b Knockdown effects on Dcp1a and Fmrp by their respective targeting siRNAs are detectable at the subcellular level. Consistent with changes at the level of protein abundance, immunofluorescence analysis shows that increased Dcp1a expression in Fmr1 knockdown is accompanied by enhanced Dcp1a puncta assembly, while compromising Dcp1a expression leads to enhanced Fmrp puncta assembly. Scale bars represent 15 μm except in zoomed panels where scale bars represent 8 μm. c Quantitative image analysis of b:The fluorescence intensities of Fmrp and Dcp1a following immunostaining of Scr, siFmr1, and siDcp1a samples were calculated and represented as box and whisker plots. Significant increases in Dcp1a puncta were observed in Fmrp knockdown and vice versa increased Fmrp puncta were observed in Dcp1a knockdown. Data were obtained from triplicate samples for each condition and graph shows scoring of at least n = 75 cells each for Scr (Scrambled), siFmr1 (Fmrp knockdown), siDcp1a (Dcp1a knockdown). Two-tailed paired Student’s t test results are indicated as *** p <0.001 regulation of Fmrp and Dcp1a not only at the level of Knockdown of Fmrp and Dcp1a on cell cycle regulators protein abundance, but also at the level of protein as- during quiescence and reactivation sembly into puncta. As knockdown of Fmrp and Dcp1a had opposing effects on MB proliferation, we evaluated the consequences of the knockdowns on expression of cell cycle regulators in Fmrp and Dcp1a play opposing roles in control of MB G0 and R3. In G0 conditions, we found nearly 5-fold in- proliferation crease in Cyclin A2 protein expression when Dcp1a was The results so far indicate differential mRNP granule pro- knocked down (Figs. 8e and S6C), consistent with the in- tein abundance and assembly in distinct cellular states: the creased proliferation observed in the MB condition. quiescent state is enriched in translational silencing/re- Fmrp knockdown, however did not affect protein ex- pressive complexes, whereas proliferating cells are pression of Cyclin A2 or Cyclin E (Figs. 8e and S6C), enriched in the classical mRNA decapping and decay consistent with the unchanged EdU incorporation in ei- complexes. Further, reducing the abundance of transla- ther knockdown in G0 (Fig. 8c). Moreover, Fmrp knock- tional repressor Fmrp by knockdown led to increased down G0 cells displayed reduced levels of mRNAs abundance and assembly of mRNA decay regulator Dcp1a encoding Cyclin A2, B1, E, D1, and ki67, but negligible and vice versa. To determine whether the differential en- change in the levels of transcripts encoding either cell richment of the decay and repressive complexes plays a cycle inhibitors (Cdkn1a/p21, Cdkn1b/p27) or myogenic role in the maintenance of a particular cellular state, we regulatory factors (MyoD, Pax7 and Myf5) (Fig. S7). By examined the phenotypes of the knockdown cells. Knock- contrast, Dcp1a knockdown increased levels of pro- down of Dcp1a in proliferating MB caused cells to prolif- proliferative transcripts including ki67 and Cyclin A2, erate more rapidly than control siRNA-treated cells, as B1, E, and ki67, and reduced levels of the anti- evidenced by a significant increase in cell number and 5- proliferative Cdk inhibitor, Cdkn1a/p21, consistent with Ethynyl-2´-deoxyuridine (EdU) incorporation (Fig. 8a). By increased proliferation. Interestingly, both MyoD and contrast, knockdown of Fmrp led to reduced EdU incorp- Pax7 transcript levels were strongly reduced. Taken to- oration (Fig. 8a), mimicking the reduced proliferative gether, the reciprocal molecular phenotypes of Fmrp -/- capacity seen in MuSC from Fmr1 mice (Fig. 3d). To- and Dcp1a knockdowns in cells in G0 were consistent gether, these results indicate that compromising the ex- with the observed reciprocal effect on proliferation. pression of key decapping and repressive/silencing mRNP Despite the altered mRNA profiles detailed above, proteins differentially affects proliferation in myoblasts. both Dcp1a and Fmrp knockdown cells were able to Dcp1a and Fmrp thus exert opposing effects on cell prolif- undergo arrest as evidenced by the absence of DNA syn- eration, possibly by targeting different transcripts for deg- thesis in suspension culture. However, whereas Dcp1a radation, translational repression, and/or sequestration. knockdown cells re-entered the cell cycle more rapidly To identify the regulatory nodes at which Fmrp and (see below), Fmrp knockdown cells were deficient in re- Dcp1a might exert their effects, we evaluated the expres- activation. These results suggest that Fmrp may nor- sion of cell cycle regulatory proteins by western blotting. mally enable reversible quiescence. Therefore, we In Dcp1a knockdown cells, Cyclin A2 protein expression determined whether Fmrp knockdown cells activated increased, consistent with enhanced proliferation (Figs. 8e other inhibitory programs, such as apoptosis or senes- and S6A). In Fmrp knockdown cells, by contrast, Cyclin cence. Flow cytometry of Fmrp knockdown cells re- A2 and Cyclin E protein levels were decreased, consistent vealed no increase in markers of apoptosis (Fig. S3B). with reduced proliferative capacity (Figs. 8e and S6A). Further, there was no increase in markers of senescence Roy et al. Skeletal Muscle (2021) 11:18 Page 20 of 28 Fig. 8 Knockdown of Fmrp and Dcp1a show opposing effects on the cell cycle. Proliferating myoblasts (MB) were treated with siRNAs (Scr, siDcp1a, siFmr1) for 18 h and either induced to enter G0 for 48 h or induced to differentiate for 48 h. For reactivation, G0 cells were harvested and plated on dishes or coverslips for 3 h. a–d EdU incorporation in MB (a), MT (b), G0 (c), and R3 (d): Dcp1a knockdown cells show increased incorporation of EdU in MB, MT, and R3, but not in G0, while Fmrp knockdown cells show decreased incorporation. EdU assay was performed simultaneously for all the conditions. Graphs show quantification by scoring > 500 nuclei for each condition in 3 biological replicates. * p < 0.05, ** p < 0.01. Two-tailed paired Student’s t test was performed. e Consistent with EdU incorporation, Cyclin A2 and Cyclin E show altered protein expression in Dcp1a and Fmrp knockdowns. Gapdh used as internal control. Values represent the mean + SD in 3 biological replicates (Fig. S4). Yet, as referred to earlier, Fmrp knockdown Indeed, knockdown of Dcp1a and Fmrp had a cells showed poor colony-forming ability, indicating marked impact during reactivation of quiescent cells. compromised self-renewal (Fig. S3A). Taken together, At 3 h of reactivation, Dcp1a knockdown cells preco- these results suggest that the absence of Fmrp leads to ciously displayed increased EdU incorporation com- an altered G0 state that we term “aberrant quiescence”. pared with control cells (Fig. 8d). Supporting the Roy et al. Skeletal Muscle (2021) 11:18 Page 21 of 28 Fig. 9 Knockdown of Fmrp and Dcp1a show similar effects on differentiation. Proliferating myoblasts (MB) were treated with siRNAs (Scr, siDcp1a, siFmr1) for 18 h and induced to differentiate for 48 h. a Both Fmr1 and Dcp1a knockdowns show reduced number of Myogenin nuclei. Upper Panel: Immunofluorescence of Myogenin (MyoG) and Fmrp in Scr, siFmr1 and siDcp1a. Scale bars represent 35 μm except in magnified panels where scale bars represent 17 μm. Lower panel: quantification based on 3 replicates, with > 600 nuclei scored per condition. b Knockdown of either Fmrp or Dcp1a affects fusion of myoblasts as shown by reduction in fusion index. Upper panel: immunofluorescence of myosin heavy chain (Myosin HC) and Fmrp in Scr, siFmr1, and siDcp1a. Lower panel: fusion index calculated as the ratio of the number of nuclei in myotubes with 2 or more nuclei over the total number of nuclei × 100 for n = 3 biological replicates. More than 850 nuclei were counted per condition. For a and b * p < 0.05, ** p < 0.01. Two-tailed Student’s t test was performed. c Representative western blots (from 3 biological replicates) of Myogenin (MyoG) and Myosin Heavy Chain (Myosin HC) proteins in MB and MT; Gapdh is internal control. d Densitometry of western blots of Myogenin (MyoG) and Myosin Heavy Chain (Myosin HC) proteins in MB and MT in c; Gapdh is loading control. Western blot analysis decreased expression of both myogenin and myosin when either Fmrp or Dcp1a expression is reduced. Two-tailed Student’s t test was performed, ** p < 0.01, *** p < 0.001. Values represent the mean + SD in 3 biological replicates premature entry into S phase, Dcp1a knockdown cells and displayed strong suppression of Cyclin E protein showed increased expression of both Cyclin A2 and levels (Figs. 8e and S6D), indicative of poor reactiva- Cyclin E proteins (Figs. 8eand S6D).Bycontrast, tion. Together, these data are consistent with oppos- Fmrp knockdown cells showed even less EdU cells ing effects of Fmrp and Dcp1 on proliferation, and than the control at this early activation time point, suggest that Fmrp and Dcp1a modulate quiescence Roy et al. Skeletal Muscle (2021) 11:18 Page 22 of 28 Fig. 10 Model showing how cross-regulation of Dcp1a and Fmrp alters the balance of mRNA turnover and translation. The balance of Dcp1a and Fmrp are hypothesized to control the turnover and translation of different sets of transcripts in distinct cellular states (middle row). When wild- type proliferating myoblasts (MB, center) enter quiescence (G0, left), protein synthesis is repression and stalled polysomes are detected, paralleled by enrichment of the translational repressor Fmrp into prominent puncta, whereas Dcp1a puncta diminish. In contrast, differentiation (MT, right) is associated with a reduction of both Fmrp and Dcp1a puncta, suggesting a new set point for the balance of these regulators. Perturbing expression of Dcp1a (upper row) or Fmrp (lower row) has reciprocal effects on mRNP granules, and opposing phenotypic consequences. Depletion of Dcp1a leads to increased Fmrp accumulation and assembly, whereas depletion of Fmrp leads to increased Dcp1a accumulation and assembly. Dcp1a knockdown (upper row) may increase levels of proteins that enhance cell proliferation directly (via reduced mRNA turnover), and indirectly act via increasing Fmrp to reduce translation of negative cell cycle regulators. Such hyper-proliferative Dcp1a knockdown cells are resistant to induction of quiescence. Conversely, Fmrp knockdown (lower row) may increase levels of proteins that repress the cell cycle directly (via de-repressed translation), and indirectly decrease levels of proteins that positively regulate the cell cycle (via increased Dcp1a and increased turnover of transcripts). Thus, Fmrp knockdown cells show reduced cell proliferation and Dcp1a knockdown cells show increased proliferation. The observations that normal induction of quiescence leads to increased Fmrp accumulation, whereas forced suppression of Fmrp also decreases proliferation, suggest that a threshold of Fmrp accumulation/assembly is required to balance between proliferation and quiescence. The observation that depletion of either Dcp1a or Fmrp leads to compromised differentiation may be explained by altered net translation of different sets of pro- and anti-myogenic target transcripts entry/exit potentially by targeting stability/utilization correlated with a pronounced suppression of Myogenin of cyclin transcripts. protein in the same sample, consistent with translation suppressive function of Fmrp. However, maintenance of Knockdown of either Fmrp or Dcp1a compromises the knockdown cells in differentiation conditions myogenic differentiation showed that loss of either Fmrp and Dcp1a negatively af- To assess the effects of depletion of Fmrp and Dcp1a on fected differentiation as evidenced by reduced Myogenin myogenesis, knockdown myoblasts were induced to dif- protein abundance, decreased frequency of Myogenin ferentiate for 2 days. As in proliferative conditions, nuclei, reduced Myosin Heavy Chain protein expression Dcp1a knockdown in low serum conditions also led to and significantly reduced fusion index (Figs. 9a–d and sustained EdU incorporation with a corresponding in- S6E). Taken together, these results indicate that despite crease in Cyclin A2 protein (Fig, 8b, e and S6B). By con- their opposing effects on the cell cycle, optimal levels of trast, Fmrp knockdown lead to negligible EdU both Dcp1a and Fmrp are required for myogenesis. incorporation accompanied by drastic reduction in Cyc- In summary, our data support a model (Fig. 10), where lin A2 protein compared with control. Notably, the Fmrp and Dcp1a reciprocally regulate each other at the cross-regulation of Fmrp by Dcp1a knockdown (see pre- level of protein abundance and granule assembly, differ- vious section) was most pronounced in myotubes, and entially regulate the expression of cell cycle and Roy et al. Skeletal Muscle (2021) 11:18 Page 23 of 28 myogenic proteins and thereby play critical and oppos- cells enter a suppressed state that is poised for reactiva- ing roles in the transitions between proliferation and re- tion. Strikingly, proteins involved in mRNA degradation versible quiescence, ultimately leading to compromised are enriched in MB, while G0 cells are enriched in pro- differentiation. teins involved in mRNA storage/suppression of transla- tion. In particular, quiescence is characterized by Discussion reduced expression of initiation factors, low rates of pro- In this study, we show that components of mRNP gran- tein synthesis, and potentially, stalled polysomes. ules regulate MuSC proliferation and differentiation Assembly of mRNP components into granules also dif- in vitro and myogenesis in vivo, likely through changes fers between G0 and MT. mRNP granules are assembled in the translation and turnover of mRNAs encoding key around distinct transcripts and modulate their function- regulators of MuSC dynamics. ality. These mRNP-associated transcripts may either be degraded, or remain in a stable, untranslated state, Quiescent cells display distinct mRNP complexes where the composition of a particular mRNP complex Non-dividing cells are well-known to exhibit reduced determines the fate of individual transcripts. Our study macromolecular metabolism. Here, we show that muscle reveals that in culture, mRNP granules containing dec- cells in two distinct states of cell cycle arrest elaborate apping proteins of the classical decay complex (Dcp1a, distinct mRNP granule protein expression, correlating Pat1, Edc4) (Table 3) are enriched in MB, suggesting with global protein synthesis. When MB enter perman- that ‘stockpiling’ of inactive transcripts in quiescence as ent arrest associated with differentiation to MT, robust during embryogenesis [50], may facilitate cell cycle reen- levels of protein synthesis sustain tissue-specific func- try when translation resumes. Notably, during G0, trans- tions. However, in reversible arrest (G0), which is typical lationally repressive complexes (Fmrp+) dominate, of adult stem cells, protein synthesis is restricted, and consistent with the enrichment of Fmrp+ storage Table 3 Details of antibodies used in this study Antibody Species Company Catalog # Dilution for WB Dilution for IFA Argonaute2 Rabbit CST C34C6 1 in 1000 1 in 200 Cyclin A2 Rabbit Abcam ab181591 1 in 2000 Cyclin E Rabbit Abcam ab71535 1 in 1000 Dcp1a Mouse Santa Cruz sc100706 1 in 200 1 in 50 Edc3 Mouse Santa Cruz sc365024 1 in 200 1 in 50 Edc4 Rabbit Santa Cruz ab72408 1 in 200 1 in 50 eIF4E Mouse Santa Cruz sc271480 1 in 200 1 in 50 Eif4g Goat Santa Cruz sc9602 1 in 200 1 in 50 Fmrp Rabbit Sigma 4055 1 in 1000 1 in 200 FxR1 Goat Abcam ab51970 1 in 1000 Gapdh Mouse Abcam ab8245 1 in 10000 GFP Chicken Abcam ab13970 1 in 300 GW182 Mouse Santa Cruz sc56314 1 in 200 1 in 50 γH2AX (Ser139) Mouse Upstate 05636 1 in 200 MyoD Mouse Dako M3512 1 in 250 1 in 200 MyoG Mouse Santa Cruz sc12732 1 in 250 1 in 250 Myosin Heavy Chain Mouse Hybridoma A4.1025 1 in 1 1 in 1 Pabp1 Rabbit Abcam ab21060 1 in 50 Pax7 Mouse Aviva ARP30947_P050 1in 100 RLP0 Rabbit Abcam ab101279 1 in 1000 Tia1 Mouse Santa Cruz sc166247 1 in 200 1 in 50 TiaR Goat Santa Cruz sc1749 1 in 200 1 in 50 Xrn1 Rabbit Sigma sab4200028 1 in 1000 1 in 200 Roy et al. Skeletal Muscle (2021) 11:18 Page 24 of 28 granules in quiescent muscle stem cells in vivo [17] (Fig. Dcp1a protein abundance and increased formation of 1). Our findings confirm the recent report [23] that Dcp1a puncta in Fmrp knockdown cells. Fmrp is required for MuSC function in vivo. By contrast, Dcp1a knockdown enhanced S phase mRNP puncta are thought to represent sites where the entry, and enhanced mRNA levels of positive regulators mRNP granule proteins exert their function [41, 51]. of progression – Cyclins A2, B2, and D1. These changes The increased abundance of Fmrp puncta in G0 may may be a direct effect of reduced cyclin mRNA turnover. suggest a role either in the entry into or maintenance of However, given the concomitant increase in Fmrp ex- quiescence. By contrast, the reduced Dcp1a puncta pression and puncta, indirect effects of Dcp1a knock- would suggest that Dcp1a either opposes or is not im- down on negative regulators of the cell cycle cannot be portant for quiescence. As discussed in detail below, the ruled out. For example, reduced translation of a poten- functional data is in apparent contradiction with this in- tial Fmrp target such as Cdkn1a/p21 would synergize terpretation: knocking down Fmrp expression (leading with increased cyclin mRNA expression to enhance S to lower Fmrp puncta accumulation) slows the cell cycle, phase entry. The identity of direct and indirect targets of whereas knocking down Dcp1a hastens the cell cycle. Fmrp and Dcp1a in different cellular states are currently When Fmrp expression is compromised, the cells enter not known, and would likely resolve this conundrum. into an aberrant quiescence, from which they are unable The opposing phenotypes of the Dcp1a and Fmrp to exit. The aberrant quiescent state is not accompanied knockdown in MB are consistent with the opposing by increased cell death or activation of senescence roles played by these proteins in regulation of the cell markers, but does exhibit compromised clonogenicity cycle. These phenotypic changes are sustained in both (self-renewal), and warrants further investigation. A pos- quiescent as well as reactivated conditions. Specifically, sible explanation is that Fmrp plays a role in the transla- increased pro-proliferative transcripts and decreased cell tional pausing we observed in the primed or poised cycle inhibitor transcripts are observed in Dcp1a knock- quiescent state, as first reported by Crist et.al., [17]. If in down cells in G0, and the converse in Fmrp knockdown absence of Fmrp its target mRNAs are continuously cells. However, the increased Cyclin A2 protein expres- translated, the cell might be unable to leave quiescence. sion in Dcp1a knockdown is not accompanied by in- Another possibility is that, as Dcp1a expression and as- creased EdU incorporation in G0, suggesting that other sembly are enhanced in Fmrp knockdown cells, tran- elements act to maintain quiescence. scripts that would normally be stabilized in a The effects of Fmrp knockdown on the activation out translationally repressed state (associated with Fmrp) of quiescence in culture were reminiscent of the re- -/- now become targets for more rapid turnover by Dcp1a. stricted proliferation of Fmr1 MuSCs, reflected by un- We hypothesize that among these destabilized tran- changed EdU incorporation and decreased Cyclin A2 scripts would be those required for the exit from quies- and Cyclin E. Fmrp may directly target cell cycle tran- cence. A detailed understanding of the direct and scripts, blocking their translation in quiescence, but sta- indirect targets of Fmrp and Dcp1a in different cellular bilizing them for mobilization during reactivation. In states is needed to resolve this issue. absence of Fmrp during cell cycle entry, these transcripts may instead be targeted for translation and turnover, by the increase in Dcp1a. The phenotype of Dcp1a knock- Reciprocal effects of Fmrp and Dcp1a on the cell down during cell cycle re-entry from G0 was similar to cycle may reflect the balance between mRNA that of cycling myoblasts: increased EdU incorporation turnover and translation in control of cell state accompanied by increased Cyclin A2 and Cyclin E ex- Perturbing Fmrp and Dcp1a expression in proliferating pression. As mRNP puncta are assembled in cell cycle cells had contrasting impacts on the cell cycle. In cycling activated myoblasts within 3 h, the Dcp1a knockdown cells, Fmrp knockdown led to an increase in Dcp1a and may accelerate proliferation via increased accumulation a simultaneous reduction in EdU incorporation, suggest- of pro-cell cycle transcripts and increased translation, ra- ing that increased nonsense-mediated decay may lead to ther than sequestration. degradation of target mRNAs, and compromise S phase All cellular states express some transcripts with ex- entry. Support for this hypothesis comes from the obser- tremely short half-lives, and likely targets of Dcp1a. The vation that expression of Cyclin E, a key positive regula- work of the Coller group [52] has shown that genome- tor of the G1/S transition is suppressed in the Fmrp wide half-lives of transcripts are increased during quies- knockdown. Given the enrichment of Fmrp stalling cence in human fibroblasts, but regulators of this change complexes and the severe translational block in G0, a re- were not identified. The targets of Dcp1a in different quirement for Fmrp in sustaining expression of Cyclin E cellular states are likely to be numerous and context- may appear paradoxical. However, it is also possible that specific, as suggested by studies in oocytes and embryos diminished expression of Cyclin E reflects increased (Ma et al (2013); Eulalio et al, (2007) [53, 54]. In our Roy et al. Skeletal Muscle (2021) 11:18 Page 25 of 28 study, we show that transcripts of cyclins A2, B1, D1, E observations may be explained by a model wherein the and Ki67 are increased when Dcp1a expression is com- translation repression normally effected by Fmrp on tar- promised in G0, providing possible direct targets. The get mRNAs in proliferating myoblasts would be lifted in model in Fig. 10 outlines the possible targets in other the Fmrp knockdown, with consequent increase in states, suggesting reduced turnover and translation of Dcp1a-associated NMD complex resulting in possible pro-proliferative transcripts during quiescence and pro- degradation of transcripts including cyclins. The knock- myogenic differentiation respectively in Dcp1a-depleted down of Dcp1a could also lead to a decrease in the cells. By contrast, non-translating mRNAs held in re- ARE-mediated decay pathway [54, 55] leading to in- pressive mRNP granule complexes containing Fmrp creased half-life of cyclin and cytokine transcripts, po- would be expected to become targets for increased turn- tentiating cell cycle progression by preventing entry into over under conditions where Dcp1a is induced by G0 [26, 28]. Conceivably, altering the flux of different knockdown of Fmrp. Considering the compromised pro- transcripts through distinct puncta could alter the pro- liferation of Fmrp knockdown cells, we hypothesize that file of proteins synthesized, impacting proliferation. pro-proliferative transcripts are likely targets for rapid The molecular mechanism for reciprocal regulation turnover, as depicted in the model in Fig. 10. Indeed, that we observe in Fmrp and Dcp1a knockdown muscle our finding that mRNAs for Cyclins D, E A, and B all cells may involve direct mRNA binding by each protein, show decreased levels in the Fmrp knockdown and in- or may be mediated by indirect regulation of upstream creased levels in the Dcp1a knockdown (Fig. S7) is con- regulators. However, as knockdown of Fmrp and Dcp1a sistent with this scenario. each have broad phenotypic consequences for the cell cycle and differentiation, it is also possible that the al- tered levels of each protein (in the context of knock- Both Fmrp and Dcp1a are necessary for normal down of the other) are associated with the altered differentiation cellular state. At present, we cannot distinguish whether Whereas Fmrp and Dcp1a have opposing effects on pro- the mechanisms involving mRNP granules we describe liferation consistent with their opposing functions in are directly responsible for regulating MuSC fate, or a mRNA turnover vs. translation, differentiation is sup- consequence of signaling that affects global translation pressed when either Fmrp or Dcp1a are perturbed. Al- and that consequently impacts mRNP granules. though the direct targets are currently unknown, the Overall, these observations indicate that both Dcp1a mechanisms by which these two regulators affect myo- and Fmrp may play a role in the assembly of mRNP genesis are likely to differ. As Fmrp knockdown leads to complexes, and that individually their knockdown affects reduced proliferative capacity, reduced differentiation the expression of transcripts encoding other mRNP pro- may reflect the reduced number of cells available for teins. Dcp1a knockdown had more pronounced effects myogenic commitment. It has been reported [23] that on transcript abundance than Fmrp knockdown, consist- -/- Fmr1 MuSCs show lower accumulation of MyoD and ent with the expected differential effects of mRNA deg- Myf5 proteins through translational silencing, delaying radation versus translational stalling. Thus, Dcp1a entry into the differentiation program. Our study sug- knockdown, by altering transcript levels of Fmr1 a key gests that this effect could be at multiple regulatory mRNP player in G0-inducing conditions, may alter the nodes where Fmrp either directly or indirectly partici- equilibrium between mRNA decay and sequestration re- pates in decisions regarding cell fate. The effect of Dcp1a quired for achieving and maintaining the quiescent state. knockdown on differentiation is consistent with the ob- Our studies point to integrative mechanisms regulating served increase in proliferation, the antagonistic nature a critical balance between the mRNA decay and transla- of these programs being well reported. At a mechanistic tional repression, which enables expression of cell cycle level, the loss of differentiation potential may reflect the (and other) regulators that control proliferation, strong reduction of Myogenin protein. quiescence or differentiation. In summary, our results support a model where distinct mRNP constellations Cross regulation of mRNP granule components revealed characterize different cellular states and suggest that re- by knockdown analysis modeling these complexes may contribute to the transi- Knockdown of Fmrp resulted in an increase in Dcp1a tions between states. puncta, and knockdown of Dcp1a led to an increase of Abbreviations Fmrp in puncta (Fig. 7b) suggesting a reciprocal balance Ago2: Argonaute 2; CD45: Cluster of Differentiation 45; Dcp1a: Decapping between mRNA decay and translational repression. Spe- Protein 1a; Edc3: Enhancer of Decapping 3; Edc4: Enhancer of Decapping 4; EdU: 5-Ethynyl-2´-deoxyuridine; eIF-4E: Eukaryotic initiation factor 4E; eIF- cifically, our results point to a regulatory loop where 4F: Eukaryotic initiation factor 4F; Fmrp: Fragile X Mental Retardation Protein; Fmrp negatively regulates Dcp1a function and Dcp1a Ki67: Marker of Proliferation; LSm4: U6 snRNA-associated Sm-like protein negatively regulates Fmrp function (Fig. 10). Our LSm4; mRNP: Messenger Ribonucleoprotein; MuSC: Muscle Stem Cells; Roy et al. Skeletal Muscle (2021) 11:18 Page 26 of 28 Myf5: Myogenic factor 5; MyoD: Myogenic Determination Protein 1; OPP: O- Myogenin is exclusively expressed by MT, Cyclin D1 is enriched in MB, Propargyl Puromycin; P0: 60S acidic ribosomal protein P0; Pat1: DNA and G0 cells lack both proteins, confirming their quiescence by the ab- topoisomerase 2-associated protein PAT1; Pax7 : Paired box 7; PFA: Para- sence of both proliferation and differentiation programs. (B) Global Tran- formaldehyde; PB: Processing bodies; Scr: Scrambled; SG: Stress granules; script status: The global transcript levels of Cyclin D1, p27, MyoD1, Myf5, TA: Tibialis anterior; TIA1: Tia1 cytotoxic granule-associated RNA binding pro- Gapdh in MB, G0, MT were analysed using data sets from Venugopal tein; VCAM-1: Vascular cell adhesion molecule 1 et.al., [49] 2020. GEO database (Series GSE110742). *p < 0.05. **p < 0.01, ***p <0.001. Figure S6. Dcp1a and Fmrp Knockdowns have opposing ef- fects on cell proliferation but compromise differentiation. This data repre- sents quantification of western blots shown in Figure 8. (A to D): Supplementary Information Densitometry of western blots of Dcp1a, Fmrp , Cyclin A2 , Cyclin E pro- The online version contains supplementary material available at https://doi. teins normalized with Gapdh as internal control in MB , MT, G0 and R3 org/10.1186/s13395-021-00270-9. (E): Western blots of Myogenin and Myosin Heavy chain proteins in G0 and R3 with Gapdh as internal control. All bar graphs represent mean ± Additional file 1: Table 1. Bio-informatic analysis of transcripts encod- sd from n ≥ 3 ,Two tailed paired Student’s t-test is indicated is indicated ing mRNP components. To assess whether changes in expression of as *p < 0.05. *p < 0.01, ***p <0.001. Figure S7. Altered expression of cell mRNP proteins resulted from changes in expression of their mRNAs, we cycle and myogenic transcripts in knockdown cells held in G0-inducing con- used the recent RNA seq analysis derived from muscle satellite cells fixed ditions. Cells were transfected with siFmr1 and siDcp1a pools for 18 by perfusion of adult mice (to prevent cell activation that results from dis- hours, then placed in suspension culture and 48 hours later RNA was iso- ruption of the niche during isolation [42]. These fixed satellite cells are lated for qRT-PCR of Dcp1a, Fmr1, Ki67, Cyclin A2, Cyclin D1, Cyclin E1, thought to more accurately represent the quiescent (G0 state) and have Cyclin B, p27, p21, MyoD1, and Myf 5 Loss of Dcp1a leads to increased a transcriptome profile distinct from MuSC isolated without fixation, abundance of transcripts encoding positive regulators of the cell cycle which are now understood to represent cells in an early activation state. (Ki67, Cyclins), along with suppression of Cdk inhibitor p21 mRNA levels, Activated satellite cells (ASC) represent proliferating primary myoblasts 2.5 consistent with increased EdU incorporation. Transcripts encoding Pax7 days post isolation from the animal. Transcripts encoding P body genes and MyoD were suppressed. Knockdown of Fmr1 leads to reduced abun- were selected from the RNAseq data and grouped according to their dance of Cyclin A2, B and D1 mRNAs, consistent with decreased EdU in- function as outlined [43, 44]. We calculated fold changes from FPKM corporation. Gapdh was used as internal control and normalized to Scr values (Fragments Per Kilobase of transcript per Million mapped reads) G0 condition. All bar graphs represent mean ± sd from n =3. Two tailed RNA seq data comparing fixed (quiescent) satellite cells and activated sat- paired Student’s t-test is indicated is indicated as *p < 0.05. **p < 0.01, ellite cells [44] and used a cut-off of 1.5 +/- (for up regulation and down ***p <0.001. regulation). False Discovery Rate approach: Two stage step-up method of Benjamini, Krieger and Yekutieli was used and 10% FDR was set up for generating p values for the analysis. Figure S1. Differential association of decay complex proteins in different cellular states. Immuno-staining of Acknowledgements Dcp1a/Edc4/Pat1 (left) and Dcp1a/Ago2 (right) in muscle cells in culture: We thank Debarya Saha (CCMB) for help with global transcript analysis and quiescent (G0), 3 hr reactivated (R3), proliferative (MB), and differentiated Louise Moyle (KCL) for help with single myofibers. We gratefully (MT). Blue arrows indicate co-localization of Dcp1a/Edc4/Pat1 in puncta. acknowledge Colin Crist (McGill Univ.) for detailed discussions and sharing Red arrows indicate co-localization of Dcp1a/Ago2 in puncta. Note the data before publication, U Varshney (IISc), R Muddashetty and Vishal Tiwari absence of Dcp1a or Pat1 puncta in G0, and the rapid reassembly in R3. (InStem) for help with polysome profiling and S Chattarji (NCBS) for access to Also note prominent nuclear staining for Edc4 in G0. Figure S2. (A) Sup- Fmr1 knockout mouse tissue used in this study. We thank J Joseph (NCCS) plementary to Figure 4A Additional representative immunofluorescence for Dcp1a antibody. We are grateful for access to flow cytometry and images showing Fmrp (green) and Dcp1a (red) puncta in G0, MB and MT, imaging facilities (CIFF at NCBS-InStem and AIF at CCMB), as well as use of as well as cells reactivated for 3 hr from G0 (R3). Arrows indicate promin- the National Mouse Resource animal facility at NCBS-InStem and the Labora- ent puncta. (B). Corrected Mean Fluorescence intensities (CMI) of Fmrp tory Animal Facility at CCMB. and Dcp1a respectively in MB, G0, R3 and MT. For quantification, more than 3 cells per group was used and CMI intensities from more than 12 puncta were analysed. Corrected mean intensity was calculated using Authors’ contributions CMI= Total intensity of signal – (Area of signal x Mean background sig- JD, NR, FPS, MP, and SS conceived and designed the study. NR, FPS, MP, SG, nal). For quantification, more than 3 cells per group was used and MFI in- HG, AA, MR, and SS performed experiments and collected data. SMH and PZ tensities from more than 10 puncta was analysed. Figure S3. Knockdown supported and supervised isolation and analysis of MuSC on single mouse of Fmrp leads to reduced cell renewability and is not accompanied by apop- myofibers by FPS. JD, NR, MP, FPS, SMH, and SS wrote the paper. The tosis. (A) Colony formation assay shows that reduced EdU incorporation authors read and approved the final manuscript. in Fmrp knockdown cells correlates with compromised self-renewal. Bar graph represents mean ± sd from n=3 biological replicates. Two tailed paired Student’s t-test is indicated as ***p<0.001. (B) Proliferating myo- Funding blasts (MB) were treated with siRNAs (Scr, siFmr1) for 18 hr and harvested This work was supported by postdoctoral fellowships from the Govt. of India at 24 hrs for FACS analysis of 10,000 cells performed after staining for Department of Biotechnology (DBT) to NR, FPS and MP, graduate fellowships apoptosis markers. X-axis represents Annexin V and Y-axis represents pro- from the Council of Scientific and Industrial Research to MR, HG, and SS, and pidium iodide. Upper Panel: Flow cytometric profile for MB. Lower Panel: from the Tata Institute of Fundamental Research to AA, and grants from the Quantification of FACS plots shows that Fmrp knockdown cells do not Indo-Australia Biotechnology Fund (DBT) and the Indo-Danish Strategic Re- undergo apoptosis. n=3, mean ± sd. Figure S4. Knockdown of Fmrp is search Fund (DBT) to JD. Animal experiments were carried out at the Na- not accompanied by senescence. (A). SA β-galactosidase assay performed tional Mouse Resource in NCBS-InStem (partially funded by DBT grant BT/ in MB cells treated with siRNAs (Scr, siFmr1) for 24 h or reactivated from PR5981/MED/31/181/2012). SMH is a Medical Research Council Scientist with quiescence for proliferation for 24 hrs (R24) does not show any significant Programme Grant (G1001029 and MR/N021231/1) support. The lab of P.S.Z. is difference in X-gal staining between control and Fmr1 Knock down cells. supported in this work by grants from the Medical Research Council (MR/ (B). Analysis of DNA damage-induced foci of γH2AX in cells reactivated P023215/1 and MR/S002472/1). The collaborative work was supported by a from quiescence for 24h (R24) does not reveal any increase in Fmr1 BBSRC Partnership Award to King’s College London and InStem. knock down cells. (C). qRT-PCR analysis for p21 did not reveal any signifi- cant change in Fmr1 knocked down condition in MB, G0 or R24. All bar graphs represent mean ± sd from n =2. Two tailed paired Student’s t-test Availability of data and materials is indicated as ***p <0.001. Figure S5. (A): Phenotyping of 3 cellular condi- All mouse strains, cell lines, antibodies, and siRNAs are available tions: A replicate blot is shown of the data depicted in Figure 4C. commercially. Roy et al. Skeletal Muscle (2021) 11:18 Page 27 of 28 Declarations 17. Crist CG, Montarras D, Buckingham M. Short article muscle satellite cells are primed for myogenesis but maintain quiescence with sequestration of Myf5 Ethics approval and consent to participate mRNA targeted by microRNA-31 in mRNP granules. STEM. 2012;11(1):118–26 All experiments were performed under appropriate institutional approvals. https://doi.org/10.1016/j.stem.2012.03.011. 18. Vogler TO, Wheeler JR, Nguyen D, Hughes MP, Britson KA, Lester E, et al. TDP-43 and RNA form amyloid-like myo-granules in regenerating muscle. Consent for publication Nature. 2018; https://doi.org/10.1038/s41586-018-0665-2. All authors have given consent for submission of this manuscript. 19. Ciaccio C, Fontana L, Milani D, Tabano S, Miozzo M, Esposito S. Fragile X syndrome: a review of clinical and molecular diagnoses. Ital J Pediatr. 2017; Competing interests 43(1):39 http://ijponline.biomedcentral.com/articles/10.1186/s13052-017-03 All authors declare no competing interests. 55-y. 20. Mazroui R. Trapping of messenger RNA by Fragile X mental retardation Author details protein into cytoplasmic granules induces translation repression. Hum Mol Institute for Stem Cell Science and Regenerative Medicine, Bangalore, India. Genet. 2002;11(24):3007–17. 2 3 Centre for Cellular and Molecular Biology, Hyderabad, India. National 21. Muddashetty RS, Kelić S, Gross C, Xu M, Bassell GJ. Dysregulated Center for Biological Sciences, Bangalore, India. Manipal Academy of Higher metabotropic glutamate receptor-dependent translation of AMPA receptor Education, Manipal, India. King’s College London, Randall Centre for Cell & and postsynaptic density-95 mRNAs at synapses in a mouse model of Molecular Biophysics, New Hunt’s House, Guy’s Campus, London, UK. fragile X syndrome. J Neurosci. 2007;27(20):5338–48. 22. Khandjian EW, Bardoni B, Corbin F, Sittler A, Giroux S, Heitz D, et al. Novel Received: 4 November 2020 Accepted: 28 April 2021 isoforms of the fragile X related protein FXR1P are expressed during myogenesis. 1998;7(13):2121–8. 23. Fujita R, Zismanov V, Jacob JM, Jamet S, Asiev K, Crist C. Fragile X mental retardation protein regulates skeletal muscle stem cell activity by regulating References the stability of Myf5 mRNA. Skelet Muscle. 2017;7(1):1–10. 1. Subramaniam S, Sreenivas P, Cheedipudi S, Reddy VR, Shashidhara LS, 24. Vasudevan S, Steitz JA. AU-rich-element-mediated upregulation of Chilukoti RK, et al. Distinct transcriptional networks in quiescent myoblasts: translation by FXR1 and Argonaute 2. Cell. 2007;128(6):1105–18. a role for Wnt signaling in reversible vs. irreversible arrest. PLoS One. 2013; 25. Brooks SA, Blackshear PJ. Tristetraprolin (TTP): interactions with mRNA and 8(6):e65097 https://doi.org/10.1371/journal.pone.0065097. proteins, and current thoughts on mechanisms of action. Biochim Biophys 2. Sebastian S, Sreenivas P, Sambasivan R, Cheedipudi S, Kandalla P, Pavlath Acta. 2013;1829(6–7):666–79. GK, et al. MLL5, a trithorax homolog, indirectly regulates H3K4 methylation, 26. Sachidanandan C, Sambasivan R, Dhawan J. Tristetraprolin and LPS- represses cyclin A2 expression, and promotes myogenic differentiation. Proc inducible CXC chemokine are rapidly induced in presumptive satellite cells Natl Acad Sci U S A. 2009;106(12):4719–24. in response to skeletal muscle injury. J Cell Sci. 2002;115(13):2701–12. 3. Cheedipudi S, Puri D, Saleh A, Gala HP, Rumman M, Pillai MS, et al. A fine 27. Hausburg MA, Doles JD, Clement SL, Cadwallader AB, Hall MN, Blackshear balance: epigenetic control of cellular quiescence by the tumor suppressor PJ, et al. Post-transcriptional regulation of satellite cell quiescence by TTP- PRDM2/RIZ at a bivalent domain in the cyclin a gene. Nucleic Acids Res. mediated mRNA decay. Elife. 2015;2015(4):1–18. 2015;43(13):6236–56. 28. Rodgers JT, King KY, Brett JO, Cromie MJ, Charville GW, Maguire KK. 4. Cheung TH, Rando TA. Molecular regulation of stem cell quiescence. Nat mTORC1 controls the adaptive transition of quiescent stem cells from G0 to Rev Mol Cell Biol. 2013;14(6):329–40. GAlert. Nature. 2014;510(7505):393–6 https://doi.org/10.1038/nature13255. 5. Dhawan J, Laxman S. Decoding the stem cell quiescence cycle - lessons 29. Sambasivan R, Gayraud-Morel B, Dumas G, Cimper C, Paisant S, Kelly RG, et from yeast for regenerative biology. J Cell Sci. 2015;128(24):4467–74. al. Distinct regulatory cascades govern extraocular and pharyngeal arch 6. Liu L, Cheung TH, Charville GW, Hurgo BMC, Leavitt T, Shih J, et al. muscle progenitor cell fates. Dev Cell. 2009;16(6):810–21. Chromatin modifications as determinants of muscle stem cell quiescence 30. Yan QJ, Asafo-Adjei PK, Arnold HM, Brown RE, Bauchwitz RP. A phenotypic and chronological aging. Cell Rep. 2013;4(1):189–204 https://doi.org/10.101 and molecular characterization of the fmr1-tm1Cgr fragile X mouse. Genes 6/j.celrep.2013.05.043. Brain Behav. 2004;3(6):337–59. 7. Dhawan J, Rando TA. Stem cells in postnatal myogenesis: molecular 31. Fukada SI, Higuchi S, Segawa M, Koda KI, Yamamoto Y, Tsujikawa K, et al. mechanisms of satellite cell quiescence, activation and replenishment. Purification and cell-surface marker characterization of quiescent satellite Trends Cell Biol. 2005;15(12):666–73 https://doi.org/10.1016/j.tcb.2005.10.007. cells from murine skeletal muscle by a novel monoclonal antibody. Exp Cell 8. Comai G, Tajbakhsh S. Molecular and cellular regulation of skeletal Res. 2004;296(2):245–55. myogenesis. Curr Top Dev Biol. 2014;110:1–73 https://doi.org/10.1016/B978- 32. Benecke BJ, Ben-Ze’ev A, Penman S. The regulation of RNA metabolism in 0-12-405943-6.00001-4 Elsevier Inc. suspended and reattached anchorage-dependent 3 T6 fibroblasts. J Cell 9. Zismanov V, Chichkov V, Colangelo V, Jamet S, Wang S, Syme A, et al. Physiol. 1980;103(2):247–54. Phosphorylation of eIF2?? Is a translational control mechanism regulating 33. Arora R, Rumman M, Venugopal N, Gala H, Dhawan J. Mimicking muscle muscle stem cell quiescence and self-renewal. Cell Stem Cell. 2016;18(1):79– stem cell quiescence in culture: methods for synchronization in reversible arrest. Methods Mol Biol. 2017;1556:283–302. 10. Buchan JR, Parker R. Eukaryotic stress granules : the ins and out of 34. Milasincic DJ, Dhawan J, Farmer SR. Anchorage-dependent control of translation. What are stress granules? Mol Cell. 2009;36(6):932. muscle-specific gene expression in C2C12 mouse myoblasts. In Vitro Cell 11. Buchan JR. mRNP granules. Assembly, function, and connections with Dev Biol Anim. 1996;32(2):90–9. disease. RNA Biol. 2014;11(8):1019–30. 35. Faye MD, Graber TE, Holcik M. Assessment of selective mRNA translation in 12. Ramaswami M, Taylor JP, Parker R. Altered ribostasis: RNA-protein granules mammalian cells by polysome profiling. J Vis Exp. 2014;92:1–8. in degenerative disorders. Cell. 2013;154:727–36 https://doi.org/10.1016/j. 36. Parker R, Sheth U. P bodies and the control of mRNA translation and cell.2013.07.038 Elsevier Inc. degradation. Mol Cell. 2007;25:635–46. 13. Brengues M. Movement of eukaryotic mRNAs between polysomes and 37. Vasudevan S, Tong Y, Steitz JA. Switching from repression to activation: cytoplasmic processing bodies. Science. 2005;310(5747):486–9 http://www. microRNAs can up-regulate translation. Science. 2007;318(5858):1931–4. sciencemag.org/cgi/doi/10.1126/science.1115791. 14. Beckham CJ, Parker R. P Bodies, stress granules, and viral life cycles. Cell 38. Darnell JC, Van Driesche SJ, Zhang C, Hung KYS, Mele A, Fraser CE, et al. Host Microbe. 2008;3(4):206–12. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function 15. Decker CJ, Parker R. P-Bodies and stress granules : possible roles in the and autism. Cell. 2011;146(2):247–61 https://doi.org/10.1016/j.cell.2011.06. control of translation and mRNA degradation. 2012;4(9):a012286 https://doi. 013. org/10.1101/cshperspect.a012286. 39. Luo Y, Shan G, Guo W, Smrt RD, Johnson EB, Li X, et al. Fragile x mental 16. Aziz A, Liu QC, Dilworth FJ. Regulating a master regulator: establishing retardation protein regulates proliferation and differentiation of adult neural tissue-specific gene expression in skeletal muscle. Epigenetics. 2010;5(8): stem/progenitor cells. PLoS Genet. 2010;6(4):e1000898 https://doi.org/10.13 691–5. 71/journal.pgen.1000898. Roy et al. Skeletal Muscle (2021) 11:18 Page 28 of 28 40. Laggerbauer B, Ostareck D, Keidel EM, Ostareck-Lederer A, Fischer U. Evidence that fragile X mental retardation protein is a negative regulator of translation. Hum Mol Genet. 2001;10(4):329–38. 41. Kedersha N, Stoecklin G, Ayodele M, Yacono P, Lykke-Andersen J, Fitzler MJ, et al. Stress granules and processing bodies are dynamically linked sites of mRNP remodeling. J Cell Biol. 2005;169(6):871–84. 42. Yue L, Wan R, Luan S, Zeng W, Cheung TH. Dek modulates global intron retention during muscle stem cell quiescence exit. Dev Cell. 2020;53(6):661– 676.e6 https://doi.org/10.1016/j.devcel.2020.05.006. 43. Zheng D, Chen CYA, Shyu A. Bin. Unraveling regulation and new components of human P-bodies through a protein interaction framework and experimental validation. Rna. 2011;17(9):1619–1634. 44. Vanderweyde T, Apicco DJ, Youmans-Kidder K, Ash PEA, Cook C, Lummertz da Rocha E, et al. Interaction of tau with the RNA-binding protein TIA1 regulates tau pathophysiology and Toxicity. Cell Rep. 2016;15(7):1455–66. 45. Moon SL, Morisaki T, Khong A, Lyon K, Parker R, Stasevich TJ. Multicolour single-molecule tracking of mRNA interactions with RNP granules. Nat Cell Biol. 2019;21(2):162–8. 46. Ben-Ze’ev A, Farmer SR, Penman S. Mechanisms of regulating tubulin synthesis in cultured mammalian cells. Cell. 1979;17(2):319–25. 47. Fukada S, Uezumi A, Ikemoto M, Masuda S, Segawa M, Tanimura N, et al. Molecular signature of quiescent satellite cells in adult skeletal muscle. Stem Cells. 2007;25(10):2448–59. 48. Seale P, Sabourin LA, Girgis-Gabardo A, Mansouri A, Gruss P, Rudnicki MA. Pax7 is required for the specification of myogenic satellite cells. Cell. 2000; 102(6):777–86. 49. Venugopal N, Ghosh A, Gala H, Aloysius A, Vyas N, Dhawan J. The primary cilium dampens proliferative signaling and represses a G2/M transcriptional network in quiescent myoblasts. BMC Mol Cell Biol. 2020;21(1):25. 50. Patel PH, Barbee SA, Blankenship JT. GW-bodies and P-bodies constitute two separate pools of sequestered non-translating RNAs. PLoS One. 2016; 11(3):1–23. 51. Eulalio A, Behm-Ansmant I, Izaurralde E. P bodies: at the crossroads of post- transcriptional pathways. Nat Rev Mol Cell Biol. 2007;8(1):9–22. 52. Johnson EL, Robinson DG, Coller HA. Widespread changes in mRNA stability contribute to quiescence-specific gene expression patterns in a fibroblast model of quiescence. BMC Genomics. 2017;18(1):123 http://bmcgenomics. biomedcentral.com/articles/10.1186/s12864-017-3521-0. 53. Ma J, Flemr M, Strnad H, Svoboda P, Schultz RM. Maternally recruited DCP1A and DCP2 contribute to messenger RNA degradation during oocyte maturation and genome activation in Mouse1. Biol Reprod. 2013;88(1):1–12 https://academic.oup.com/biolreprod/article-lookup/doi/10.1095/ biolreprod.112.105312. 54. Eulalio A, Rehwinkel J, Stricker M, Huntzinger E, Yang SF, Doerks T, et al. Target-specific requirements for enhancers of decapping in miRNA- mediated gene silencing. Genes Dev. 2007;21(20):2558–70. 55. Stoecklin G, Mayo T, Anderson P. ARE-mRNA degradation requires the 5′ -3′ decay pathway. EMBO Rep. 2006;7(1):72–7. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Skeletal Muscle Springer Journals

mRNP granule proteins Fmrp and Dcp1a differentially regulate mRNP complexes to contribute to control of muscle stem cell quiescence and activation

Loading next page...
 
/lp/springer-journals/mrnp-granule-proteins-fmrp-and-dcp1a-differentially-regulate-mrnp-YDbE4Y601L
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2044-5040
DOI
10.1186/s13395-021-00270-9
Publisher site
See Article on Publisher Site

Abstract

Background: During skeletal muscle regeneration, satellite stem cells use distinct pathways to repair damaged myofibers or to self-renew by returning to quiescence. Cellular/mitotic quiescence employs mechanisms that promote a poised or primed state, including altered RNA turnover and translational repression. Here, we investigate the role of mRNP granule proteins Fragile X Mental Retardation Protein (Fmrp) and Decapping protein 1a (Dcp1a) in muscle stem cell quiescence and differentiation. Methods: Using isolated single muscle fibers from adult mice, we established differential enrichment of mRNP granule proteins including Fmrp and Dcp1a in muscle stem cells vs. myofibers. We investigated muscle tissue homeostasis in adult Fmr1-/- mice, analyzing myofiber cross-sectional area in vivo and satellite cell proliferation ex vivo. We explored the molecular mechanisms of Dcp1a and Fmrp function in quiescence, proliferation and differentiation in a C2C12 culture model. Here, we used polysome profiling, imaging and RNA/protein expression analysis to establish the abundance and assembly status of mRNP granule proteins in different cellular states, and the phenotype of knockdown cells. * Correspondence: jdhawan@ccmb.res.in; jdhawan@instem.res.in Nainita Roy, Swetha Sundar, Malini Pillai and Farah Patell-Socha contributed equally to this work. Institute for Stem Cell Science and Regenerative Medicine, Bangalore, India Centre for Cellular and Molecular Biology, Hyderabad, India Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Roy et al. Skeletal Muscle (2021) 11:18 Page 2 of 28 Results: Quiescent muscle satellite cells are enriched for puncta containing the translational repressor Fmrp, but -/- not the mRNA decay factor Dcp1a. MuSC isolated from Fmr1 mice exhibit defective proliferation, and mature myofibers show reduced cross-sectional area, suggesting a role for Fmrp in muscle homeostasis. Expression and organization of Fmrp and Dcp1a varies during primary MuSC activation on myofibers, with Fmrp puncta prominent in quiescence, but Dcp1a puncta appearing during activation/proliferation. This reciprocal expression of Fmrp and Dcp1a puncta is recapitulated in a C2C12 culture model of quiescence and activation: consistent with its role as a translational repressor, Fmrp is enriched in non-translating mRNP complexes abundant in quiescent myoblasts; Dcp1a puncta are lost in quiescence, suggesting stabilized and repressed transcripts. The function of each protein differs during proliferation; whereas Fmrp knockdown led to decreased proliferation and lower cyclin expression, Dcp1a knockdown led to increased cell proliferation and higher cyclin expression. However, knockdown of either Fmrp or Dcp1a led to compromised differentiation. We also observed cross-regulation of decay versus storage mRNP granules; knockdown of Fmrp enhances accumulation of Dcp1a puncta, whereas knockdown of Dcp1a leads to increased Fmrp in puncta. Conclusions: Taken together, our results provide evidence that the balance of mRNA turnover versus utilization is specific for distinct cellular states. Keywords: Quiescence, mRNP granule, Translational control, mRNA decay, Skeletal muscle, Myoblast, G0, Fmrp, Dcp1a, Fmr1 knockout, Muscle stem cell Background and stress granules. P-bodies are dynamic structures that During skeletal muscle regeneration, the resident muscle are enriched in proteins involved in mRNA decay (such stem cells, called satellite cells (MuSc), use distinct path- as Dcp1a, Edc4, Edc3, Lsm1-7 complex) [13, 14]. In con- ways to either enter myogenic differentiation to restore trast, stress granules may form in response to stress and functional tissue, or self-renew by returning to mitotic contain stalled translation initiation complexes (contain- quiescence to replenish the stem cell compartment. We ing Fmrp, eIF-4E, eIF-4G, Pabp, Tia-1/TiaR), and occa- have previously reported transcriptional and epigenetic sionally, 40S ribosomal subunits [15]. P-bodies and mechanisms that control the choice between these irre- stress granules share many proteins and interact with versible and reversible cell cycle arrests [1–3]. In par- each other, precluding unambiguous classification based ticular, quiescence is regulated by mechanisms that [15] on the presence or absence of individual compo- promote a poised or primed state, compatible with re- nents, and prompting the use of the inclusive nomencla- entry into the cell cycle [4]. The view of quiescence (G0) ture of “mRNP granules”. Recent reports of mRNP as an actively managed poised state, rather than an inert granule proteins in MuSC suggest a role for translational default state, is supported by several findings [5], which control and use of stored mRNA in regulation of quies- show that two major programs (the cell cycle and myo- cence and activation [16, 17] and in regeneration [18]. genesis) are held in abeyance by diverse mechanisms [6– The flux of transcripts between mRNP granules is as- 8]. In addition to transcriptional and epigenetic silencing sociated with cell state transitions, and altered aggrega- in G0, quiescent cells also exhibit translational repres- tion status is reported in neuromuscular diseases, the sion [9], but remain capable of rapid remobilization of best-studied example being fragile X syndrome (FXS). In pre-existing transcripts onto polysomes during cell cycle this devastating neuro-developmental disorder, loss of activation. the fragile X mental retardation protein (Fmrp) leads to Upon export from the nucleus, newly synthesized a spectrum of autistic features characterized by cognitive mRNAs are either rapidly assembled onto polysomes for and behavioral deficits [19]. The location of Fmrp in immediate translation or held in a non-translating com- cytoplasmic granules and its molecular function as a re- partment, bound by a variety of RNA-binding proteins pressor of activity-dependent protein translation in that control mRNA transport, localization, decay, and axons [20] suggest mechanisms by which signal- translational efficiency. RNA-binding proteins dynamic- dependent protein synthesis is required for higher-level ally coalesce, along with mature mRNAs and miRNAs, brain function [21]. Despite the nearly ubiquitous ex- into mRNP granules [10, 11]. Depending on the cellular pression of Fmrp [17], little is known of its specific role context and lineage, several kinds of mRNP granules in non-neuronal tissues, including development and re- exist [10] that, due to their distinct composition, may generation of skeletal muscle. Fmrp has been reported function differently to regulate mRNA utilization [12]. to be downregulated during muscle differentiation [22] Among mRNP granules, the best studied are P-bodies and is detected in quiescent MuSC, where it regulates Roy et al. Skeletal Muscle (2021) 11:18 Page 3 of 28 -/- MuSC function via control of the myogenic determinant using Fmr1 mice, suggesting its involvement in Myf5 mRNA [17, 23]. homeostatic and regenerative control in muscle, beyond Increasing evidence points to a role for post- its established role in neuronal function. During primary transcriptional control in quiescent cells, including qui- MuSC activation on myofibers, we found that Fmrp and escent adult MuSC. Studies in yeast and cultured fibro- Dcp1a show reciprocal expression and organization, with blasts showed that Fmrp and the related Fxr1 are Fmrp puncta prominent in quiescence, but Dcp1a important for entry into G0 [24]. In myoblasts, the entry puncta appearing during activation/proliferation. We into mitotic quiescence is associated with an induction also explored the muscle cell-intrinsic functions of Fmrp of genes encoding mRNP granule components, such as as distinct from neurological effects manifested in vivo, tristetraprolin (TTP) [25, 26], primarily involved in AU- using a cultured myoblast model of quiescence. Our re- rich element (ARE)-mediated decay, that are also re- sults suggest the existence of distinct mRNP complexes quired for MuSC regenerative function [27]. A pioneer- in different cellular states (proliferation, quiescence, and ing report by Crist et.al., [17] showed that quiescent differentiation). Specifically, whereas translational re- MuSC sequester transcripts of Myf5 in an untranslated pressive complexes containing Fmrp predominate in G0, fraction, and remobilize them onto polysomes during re- we report an enrichment of nonsense-mediated mRNA activation. Further, a general repression of protein syn- decay complexes containing the mRNA-decapping en- thesis by phosphorylation of the translation initiation zyme 1A (Dcp1a) in proliferating myoblasts, suggesting factor eIF2α is essential for maintenance of the quiescent that post-transcriptional regulatory complexes may be state [9]. However, there is little information available remodeled depending on cellular context. Functional on the composition and function of heterogeneous analysis using mRNA knockdown indicates that Fmrp mRNP granules that might regulate the quiescent state and Dcp1a play opposing roles in myogenic proliferation per se, and in particular, the relative roles of transla- and quiescence; Fmrp sustains proliferative potential, tional repression and mRNA decay in the entry into qui- whereas Dcp1a functions to restrain proliferation of escence. This balance may be important in the context myoblasts. Intriguingly, these opposing functions cross- of the global suppression of macromolecular synthesis, regulate, such that knockdown of Fmrp leads to in- in G0 cells [16]. For example, key regulators of protein creased Dcp1a expression and assembly into puncta, and synthesis such as mTOR control awakening of quiescent reciprocally, Dcp1a knockdown myoblasts show in- MuSC [28], but the coupling of mRNA utilization to creased Fmrp expression and assembly into puncta. metabolic activation has not been extensively explored However, unlike their opposing roles in proliferation, in MuSC. knockdown of either Dcp1a or Fmrp led to compro- MuSC function is intimately linked to the ability to mised myogenic differentiation. Taken together, our enter and exit quiescence. Whereas both differentiation study shows the importance of the balance between and quiescence are mitotically inactive states, muscle translational repression and mRNA decay in the regula- terminal differentiation is irreversible and requires pref- tion of quiescence, and indicates a role for distinct erential transcription and translation of tissue-specific mRNP granule proteins in regulating this equilibrium. proteins that comprise and control the specialized sarco- meric cytoskeleton [16]. By contrast, quiescence is re- Materials and methods versible and is characterized by a broad suppression of Single myofiber isolation and analysis the differentiation program and increased expression of Animal experiments were carried out in accordance with the MuSC-specific transcription factor Pax7, which are CPCSEA guidelines of the Govt. of India as approved by reversed during cell cycle activation, along with re- the Institutional Animal Ethics Committee of InStem induction of determination factors MyoD and Myf5 [26]. and CCMB, or in accordance with British law under pro- Indeed, quiescent MuSC exhibit translational control of visions of the Animals (Scientific Procedures) Act 1986, lineage determinants [27], with Myf5 transcripts held in as approved by the Ethical Review Process Committee of non-translating mRNPs [17, 23]. King’s College London. Prompted by our observation that Fmrp mRNA ex- EDL muscles were dissected from hind limbs of 2–7- pression is induced in G0 [1], in the present study we month-old mice of either sex (two C57BL/6 mice aged profiled expression of a set of mRNP proteins in muscle 8–10 weeks (Fig. 1) or six Pax7-nGFP mice (four males cells and explored their function in quiescence. We first aged 3 months) and two females aged 5 months (Fig. 2) surveyed the expression and distribution of mRNP com- [29]. Isolated muscles were digested with 400 U/ml Type plex proteins in MuSC versus myofibers in isolated sin- I collagenase (Worthington) in DMEM at 37 °C, till sin- gle muscle fiber preparations ex vivo. We report the gle fibers dissociated. All dissociated fibers were trans- enrichment of Fmrp bodies in MuSC in wild-type mice ferred into fresh DMEM medium and triturated gently and reveal a role for Fmrp in MuSC function in vivo to release individual fibers using fire-polished pasteur Roy et al. Skeletal Muscle (2021) 11:18 Page 4 of 28 Fig. 1 (See legend on next page.) Roy et al. Skeletal Muscle (2021) 11:18 Page 5 of 28 (See figure on previous page.) Fig. 1 Expression and distribution of mRNP granule proteins in isolated skeletal muscle fibers.A Schematic of an isolated myofiber (MF) depicting myonuclei (MN) and an associated muscle satellite cell (MuSC). The longitudinal striations represent orientation of the myofibrils while the cross- striations represent the A-band (A) and Z line (Z). B–D, B’–D’ Depict magnified views of the regions enclosed by brackets (dotted lines) in B–D to visualize subcellular distribution of Fmrp. Arrow heads indicate cytoplasm, double arrows indicate nucleus. Fmrp puncta are observed both in nucleus and cytoplasm of MuSC (B, B’) and as cross-striated staining in myofiber. Puncta also accumulate in a cytoplasmic domain adjacent to the MN, while MN is itself not stained (C, C’). Nuclear accumulation of Fmrp is also seen in the Pax7 MuSC nucleus (D, D’) but not in an adjacent Pax7 MN. B’, C’, and D’ represent single-channel (488) images. E–G Distribution of Fmrp puncta (green) in myofiber in a cross-striated pattern congruent with Z lines revealed by α-actinin (red). Arrows in G point to Fmrp puncta co-localizing with α-actinin striations. H–I. Secondary antibody controls (mouse and rabbit) do not show either punctate or striated background. K Distinct GW182 bodies are visible in Pax7 MuSC. Pax7 MN also show distinct perinuclear puncta (arrowheads) and significant punctate staining is observed in MF cytoplasm in a doublet striated pattern likely reflecting A-band localization. K’ Region within brackets in k magnified to show GW182 puncta in the MuSC nucleus (double arrow). L, L’ No enrichment (either nuclear or cytoplasmic) is detected of Dcp1a in MuSC nucleus (marked with the membrane marker Caveolin 1). Faint fibrillar puncta are observed in myofibers. M, M’ Xrn1 is faintly detected in MuSC, but strongly expressed in myofibres in both a longitudinal and cross-striated pattern. K’, L’, and M’ represent single channel (green) images of enlarged areas indicated by brackets in K, L, and M, respectively pipettes. Dispersed single fibers were either immediately mouse per genotype), minced, and digested in collage- fixed in in 4% paraformaldehyde (PFA) for 10 min or nase type II (Cat# LS4196 Worthington Biochemical, cultured for up to 48 h in DMEM, 10% Horse Serum, 400 U/ml final concentration) for 90 min at 37 °C with 20% Fetal Bovine Serum, and 2% Chick Embryo Extract gentle vortexing after every 15 min. The digested muscle (Sera Lab, CE-650-J) followed by fixation, washed three slurry was filtered through 40-μm nylon mesh. The sin- times with PBS, picked and placed on charged slides gle cell suspension was treated with 0.8% ammonium (Thermo-Fisher) for immunostaining. Fibers were chloride to lyse RBCs. Muscle mononuclear cells were permeabilized with 0.5% Tween-20 in PBS for 1 h, washed twice with PBS and stained with biotinylated blocked in 5% BSA in PBS 0.5% Tween-20 for 1 h. Sub- anti-VCAM-1 (BD Biosciences, Cat#553331) primary sequent steps were as for cultured cells. Samples were antibody for 30 min, washed with PBS and stained with imaged on a LSM510 Meta or LSM 880 Airy Scan Streptavidin, Alexa Fluor-488 conjugate (Invitrogen, (Zeiss). Image analysis was done using ImageJ. Cat#S-11223) and CD45-PE (BD Biosciences, Cat#553081) conjugated antibody. Cell sorting was per- Muscle histology formed on Moflo XPD cytometer using gates for the + - To determine muscle cross-sectional area, 2 male adult VCAM-1 and CD45 population. The gated cell popula- -/- mice (6–8 weeks) each for Fmr1 [30] and age-matched tion was sorted directly into growth medium for subse- WT mice were used. The TA muscle was carefully dis- quent culturing on Matrigel (BD Biosciences, sected intact, fixed for 2 h in 4% PFA at 4 °C and equili- Cat#354230) coated dishes for 6 days. brated overnight in 20% sucrose at 4 °C. The muscle tissues were mounted in OCT in cryomoulds and flash Cell culture frozen in a liquid nitrogen-cooled isopentane bath. Serial The C2C12 mouse muscle cell line was cultured in 20 μm cryosections were collected and immunolabelled DMEM supplemented with 20% FBS (with Penicillin/ with anti-laminin antibody to highlight the individual Streptomycin). To generate synchrony in G0 (reversible fiber perimeters, and imaged by confocal microscopy. arrest), cells were cultured in methylcellulose suspension For calculating the cross-sectional area (CSA) of myofi- with 20% FBS for 48 h [32–34]. To differentiate cells into bers, 7 sections were chosen at random from wild-type irreversibly arrested multinucleated myotubes, myoblasts muscle and a corresponding section selected from Fmr1 at 80% confluency were cultured in low serum media -/- muscle (from the equivalent position in the TA). The (DMEM + 2% Horse serum) for 120 h with daily medium CSA of ~ 250 myofibers was measured from confocal changes; myotubes appeared by day 2, with fusion in- images (LSM510 Meta) of sections using ImageJ soft- creasing till day 5. ware, and the mean CSA and two-tailed paired Student’s t test were performed to compare the difference between Western blot analysis the two groups. Cells were lysed in 50 mM Tris HCl, pH 8, 150 mM NaCl, 5 mM MgCl , 0.1% Nonidet P-40) supplemented Isolation of mouse muscle satellite cells with complete protease inhibitor cocktail (Roche Diag- Primary MuSCs were purified from adult mice as de- nostics, France) and incubated on ice for 30 min. Soluble scribed [31]. Briefly, all hind limb muscle groups were proteins were recovered after centrifugation at 15,000 g -/- dissected from 6-week-old WT or Fmr1 mice (1 male at 4 °C for 10 min and quantified by the BCA method. Roy et al. Skeletal Muscle (2021) 11:18 Page 6 of 28 Fig. 2 Dynamics of Fmrp and Dcp1a granules in quiescent and activated MuSCs on isolated single muscle fibers.a–c Immunofluorescence analysis of Dcp1a and Fmrp in freshly isolated EDL myofibers (0 h) and at 12, 24, and 48 h of culture. MuSCs are marked with Pax7-nGFP. a Fmrp puncta are evident in quiescent MuSCs at 0 h of culture, and become dispersed at 24 h, whereas Dcp1a puncta are not evident until 24 h of activation. The patterns of staining are consistent with reciprocal patterns of Dcp1a and Fmrp assembly into granules. b The pattern of Dcp1a in + + + - proliferating MuSCs (Pax7-nGFP EdU ) confirms the timing of MuSC activation. Dcp1a puncta are not observed in quiescent Pax7 EdU cells, but + + + - are present in activated Pax7 EdU MuSC at 24 and 48 h. c By contrast to Dcp1a, the pattern of Fmrp is punctate in quiescent MuSC (Pax7 EdU ) + + and becomes diffuse in activated Pax7 EdU MuSC at 24 and 48 h. Immunofluorescence was performed in N = 3 biological replicates using > 5 EDL fibers for each combination of antibodies in each assay. Scale bars are 10 μm, or 4.8 μm in magnified panels. Proteins were separated on a 4–12% polyacrylamide peroxidase-conjugated goat anti-mouse or anti-rabbit SDS-PAGE gel along with a pre-stained protein ladder IgG—for 45 min at room temperature. After washing in (12–250 kDa) and transferred to a nitrocellulose mem- TBST, the blots were developed using chemilumines- brane. Non-specific protein binding sites were blocked cence solutions and imaged using Image Quant (Anti- by incubation in 5% (w/v) non-fat dry milk (made in body information in Table 3). TBST), for 1 h at room temperature. The membrane was then incubated with antibodies against the different Immunofluorescence and confocal analysis mRNP granule proteins overnight at 4 °C. After washing Cells were grown on cover slips placed in 6-well plates in TBST, the blot was incubated with horseradish (5 × 10 cells per 18-mm cover slip). The next day, cells Roy et al. Skeletal Muscle (2021) 11:18 Page 7 of 28 were rinsed with ice-cold PBS and fixed with 4% PFA GAAGGGACGUUAUUUGUAU. for 10 min at room temperature followed by Non-targeting siRNA pool#1: permeabilization with 0.1% Triton X-100. The cells were UAGCGACUAAACACAUCAA, subjected to immunofluorescence staining with different UAAGGCUAUGAAGAGAUAC, antibodies overnight at 4 °C. The cells were washed with AUGUAUUGGCCUGUAUUAG, cold PBS and incubated with anti-Rabbit Alexa 488 AUGAACGUGAAUUGCUCAA. (Invitrogen A11034, 1:500) and anti-Mouse Alexa 568 Quantitative real-time RT-PCR was performed on an (Invitrogen A11037, 1:500) secondary antibodies at room ABI 7900HT thermal cycler (Applied Biosystems) using temperature for 1 h. The cells were examined by con- the SDS 2.1® ABI Prism software. cDNA was prepared focal microscopy using LSM510 (Zeiss, Germany). Co- from 1 μg total RNA using superscript II (Invitrogen) localization was quantified using ImageJ software, after and used in SYBR-Green assay (Applied Biosystems). maximizing the intensity from all the Z-stacks for a par- Each sample was isolated from three independent bio- ticular image, and analyzed by selecting the puncta as logical samples and analyzed in triplicate reactions. ROI. Image intensity was calculated using Fiji (ImageJ) Amplicons were verified by dissociation curves and se- software and corrected mean intensity (CMI = Total in- quencing. Primer sequences are listed in the Supplemen- tensity of signal − Area of signal × Mean background tary Information. Relative abundance of different signal) was calculated for more than 75 cells. All data mRNAs in Fmrp and Dcp1a knockdown cells was calcu- points were plotted in Box and whisker plot, and p value lated with reference to cells transfected with non- was calculated by two-tailed paired Student’s t test. targeting siRNA and normalized to GAPDH levels. Fold change was calculated using differences in normalized −ΔΔct RNA interference using siRNA cycle threshold value 2 . The following small interfering RNAs (siRNAs) from Primer sequences used in this study: Dharmacon, Thermo Scientific were used for the study: Gapdh - F: 5’-ATCAACCGGGAAGCCCATCAC -3’ siGENOME SMARTpool siRNAs against mouse Fmrp R: 5’- CCTTTTGGCTCCACCCTTCA- 3’ (M-045448-01-0005), mouse Dcp1a (M-065144-01- Cyclin D1- F: 5’-AAGTGCGTGCAGAAGGAGATTG 0005) and non-targeting siRNA pool #1 (D-001206-13- TG-3’ 20); each pool represents 4 distinct siRNAs targeting dif- R:5’ TCGGGCCGGATAGAGTTGTCAGT-3’ ferent sequences in the same transcript. C2C12 myo- Cyclin A2- F: 5’-TTCTGGAAGCTGACCCATTC-3’ blasts maintained in growth medium (DMEM + 20% R: 5’-GGCAAGGCACAATCTCATTT-3’ Fetal bovine serum) were transfected with the siRNAs Cyclin B1- F: 5’-ATGGACACCAACTCTGCAGCAC-3’ listed above using the Lipofectamine ® RNAiMAX Re- R: 5’-CTGTGCCAGCGTGCTGATCT-3’ agent (Invitrogen) according to the manufacturer’s in- Cyclin E1- F: 5’-TGTCCTCGCTGCTTCTGCTTTG structions. Eighteen hours post-transfection, the cells TATCAT-3’ were either induced to differentiate in low mitogen R: 5’-GGCTTTCTTTGCTTG GGCTTTGTCC-3’ medium (DMEM + 2% Horse serum), for 2 days to form Dcp1a -F:5’- CCAGCTGAAGCTCCTACCAC-3’ myotubes (MT) or were synchronized to G0 in suspen- R:5’- CTGTGGGGTCAACCTGAGTT-3’ sion cultures (1.3% methylcellulose in growth medium) Fmr1 - F: 5’-AGGCTTGGCAGGGTATGGTA -3’ [33]. siRNA-transfected cultures were harvested 48 h R:5’-TGTACGATTTGGTGGTGGTCT-3’ after induction of myogenesis or quiescence and sub- Fmr1 -F: 5’-AGAGGAGGAGGCTTCAAAGG-3’ jected to different analyses including EdU proliferation R: 5’- AGAGGAGGAGGCTTCAAAGG-3’ assay, western blotting for different cell cycle proteins Myogenin- F: 5’-TGGGCATGTAAGGTGTGTAAGA-3’ (such as Cyclins A, B, D1, and E), p27, p21, and qRT- R: 5’-ACTTTAGGCAGCCGCTGGT-3’ PCR analysis. Knockdown in these cells was confirmed Pax7 - F: 5’-CATGGTGGGCCATTTCCACT-3’ by western blotting using anti-Fmrp and anti-Dcp1a R: 5’-GGCCCGGGGCAGAACTAC-3’ antibodies (Table 3). p27 - F: 5’-TGCAGTCGCAGAACTTCGAA-3’ siRNA target sequences (smart pool of 4 siRNAs per R: 5’-ACACTCTCACGTTTGACATCTTCCT-3’ transcript): p16 - F: 5’-CGAACTCGAGGAGAGCCATC-3’ Fmrp: GAUUAUCACCUGAACUAUU, R: 5’- CGTGAACGTTGCCCATCATC-3’ GAUCUGAUGGGUUUAGCUA, Myf5 - F: 5’-CCCCACCTCCAACTGCTCTG-3’ CGUCACUGCUAUUGAUUUA, R: 5’-CCAAGCTGGACACGGAGCTT-3’ GAUCAUUCCCGAACAGAUA. MyoD1 - F: 5’-AGCGTCTCGAAGGCCTCAT-3’ Dcp1a: CAACAGCUAUGGGUCUAGA, R: 5’-AGCGCAGCTGAACAAGCTA-3’ GACAGUAGAAGAGUUAUUU, Ki67 - F: 5’-TGGAAGAGCAGGTTAGCACTGT-3’ GUAUAGAAAUGCAAGUUUG, R: 5’-CAAACTTGGGCCTTGGCTGT-3’ Roy et al. Skeletal Muscle (2021) 11:18 Page 8 of 28 EdU incorporation analysis buffer composition: 10 mM Tris-Cl (pH 7.4), 150 mM EdU incorporation was performed in muscle fibers and KCl, 10 mM MgCl , 1 mM DTT, 100 μg/ml Cyclohexi- cell cultures according to the manufacturer’s protocol mide) and samples were centrifuged at 39,000 rpm for (Invitrogen EdU assay Kit Catalog No. C10340). 1.5 h at 4 °C in a SW41 Ti rotor (Beckman Coulter, Brea, CA, USA). The density-separated lysate was analyzed in OPP incorporation assay a polysome profiler linked to a fraction collector (ISCO) OPP incorporation was performed in C2C12 cells cul- with a UA-5 UV detector. Nine fractions of 1 ml were tured for different conditions according to the manufac- collected for each gradient and were used for either pro- turer’s protocol (Invitrogen OPP assay Kit Catalog No. tein analysis by immunoblot or RNA analysis by qRT- C10456). Cells were imaged using Leica SP8 TCS. Inte- PCR. IDAQ software was used for profile generation. grated fluorescence intensities were calculated using Cell The area under each ribosomal peak (40S, 60S, 80S, and Profiler. R was used to perform Multivariate ANOVA polysomes) was calculated using Microsoft Excel and the followed by Tukey’s HSD post hoc test, and ggplot was average area of polysomes was divided by the average implemented for generating box and whisker plots. area of monosomes (80S ribosomes) for each profile in Estimating fluorescence intensities in MT involved ap- order to calculate polysome/monosome (P/M) ratio. proximations in ROI (boundary) selections for the pur- For isolation of RNA from polysome profiles [21, 35], pose of quantification, as MT are multinucleated and we pooled fractions that constituted the mRNPs (frac- overlapping. tions 1 & 2), monosomes (4 & 5), and polysomes (7 & 8) from each gradient derived from CHX- and Puro-treated Apoptosis assay MB, G0, and MT cells, and isolated RNA using TRIzol™ To determine whether Fmrp knockdown cells undergo LS Reagent (Invitrogen) according to the manufacturer’s apoptosis, we used the Invitrogen Apoptosis Kit as per instructions. The isolated RNA was resuspended in the manufacturer’s protocol (Cat no. V13245). Flow cy- equal volumes of water, quantified and checked for its tometry was performed on a BD Fortessa, using FlowJo purity by Nanodrop. Equal volumes of RNA from the software for analysis. pooled fractions were subjected to cDNA synthesis using SuperScript IV (Invitrogen) according to the manufac- Senescence assays turer’s protocol. Quantitative real-time PCR analysis was Control or Fmrp knockdown cells were evaluated for ex- performed using Power SYBR Green (Applied Biosys- pression of p16 and p21 by qRT-PCR. Activity of tems) in ABI 7900HT Thermal cycler (Applied Biosys- senescence-associated β-galactosidase (β-Gal) enzyme tems). Serial dilutions of cDNA resulting in decrease in was tested through X-gal staining assay at pH 6.0 to sup- copy number were used to generate a standard curve for press lysosomal β-gal activity. Briefly, cells were fixed each gene. C values of each dilution were plotted with 4% PFA for 10 min at room temperature, washed against the copy numbers (ln of dilution). For experi- twice with PBS, and incubated with the chromogenic mental samples, the copy number was calculated using substrate 5-bromo-4-chloro-3-indolyl-beta-d-galactopyr- C value and standard curve obtained for that mRNA anoside (X-Gal) staining solution comprising 1 mg/ml with the same set of primers. The percentage of total X-gal (in DMF), 5 mM potassium ferrocyanide, 5 mM mRNA across the gradient was calculated as follows: potassium ferricyanide and 1 mM MgCl , overnight at (copy number in specific fraction/Total copy number of 37 °C. DNA damage-induced foci of γH2AX were de- the gene across the gradient) × 100. tected by immunofluorescence as detailed in the Supple- mentary Information. Results mRNP components are differentially expressed in Polysome analysis quiescent muscle stem cells versus myofibers Fifteen million cells (MB, MT, or G0) were incubated To investigate mRNP protein distribution in muscle, we with 0.1 mg/ml of Cycloheximide for 15 min or with 0.1 used isolated murine myofibers ex vivo, complete with mg/mL of Puromycin for 2 h prior to lysis. Cells were resident MuSCs in their niche. At a subcellular level, lysed in ice-cold lysis buffer (10 mM Tris-Cl (pH 7.4), mRNP granule components are known to partition be- 150 mM KCl, 10 mM MgCl , 1 mM DTT, 100 μg/ml Cy- tween a diffuse cytoplasmic distribution and punctate TM cloheximide, 1% NP40, 1× cOmplete EDTA-free Pro- granules, where puncta represent functional complexes tease inhibitor (Cat. No. 5056489001), RNAse inhibitor of RNA and proteins [36, 37]. Immunofluorescence ana- (Cat. No. 10777-019), 6 U/ml). After incubation in cold lysis revealed that mRNP granule proteins are organized lysis buffer for 30–45 min, the lysate was spun at in puncta that were enriched in quiescent Pax7 MuSC 13000×g for 15 min at 4 °C. The supernatant was then (Fig. 1). Myofibers also showed puncta, organized in stri- loaded on to 10–45% (wt/wt) sucrose gradient (gradient ated patterns that suggest association with underlying Roy et al. Skeletal Muscle (2021) 11:18 Page 9 of 28 -/- cytoskeletal elements. Specifically, the translational re- knockout (Fmr1 ) mouse. Quantification of cross- pressor Fmrp showed punctate immunolabeling that was sectional area of muscle fibers in cryo-sections of adult highly enriched in the cytoplasm of Pax7 MuSC (Fig. tibialis anterior muscle revealed that muscle fibers in -/- 1B–D), but also associated with sarcomeres in myofibers Fmr1 muscle showed drastically reduced caliber [mean (Fig. 1E–G), with distinct non-sarcomeric enrichment in ± SD of 619 μm ± 200] compared with age-matched - 2 the cytoplasm adjacent to Pax7 myonuclei (Fig. 1C, C’). wild-type (WT) mice [mean ± SD of 1518 μm ± 438, n Interestingly, Fmrp was also located in MuSC nuclei = 250 fibers; p value < 0.0001] (Fig. 3a, b). As there is (Fig. 1B, B’,D,D’), but not in myonuclei. Another trans- some expression of Fmrp in myofibers, a direct effect of lational repressor GW182, which is involved in the Ago- this mRNP granule protein in a fiber-intrinsic mechan- miRNA pathway, showed a similar distribution to Fmrp: ism cannot be ruled out. However, we also found that + -/- - + discrete cytoplasmic puncta in Pax7 MuSC and in FACS-isolated Fmr1 CD45 VCAM-1 MuSCs (Fig. 3c) zones near myonuclei, with smaller puncta in myofibers, proliferate less compared to WT controls (Fig. 3d). The -/- arranged in a distinct pattern reflecting sarcomeric proportion of sorted MuSCs was similar in Fmr1 and organization (Fig. 1 K). Thus, proteins implicated in WT (Fig. 3c), by contrast to an earlier report by Fujita +- translational repression are located in mRNP granules et al. [23], who reported that the number of Pax7 - -/- clearly evident in quiescent MuSC. MyoD cells on single fibers were lower in Fmr1 mice. To determine the distribution of proteins involved in This difference may reflect differences in the methods mRNA turnover, we examined expression of key regula- and markers used to identify MuSCs, as well as the tors of mRNA, the decapping enhancer Dcp1a and the numbers of cells analyzed in the two studies. Further, we 5’-3’ exoribonuclease Xrn1 [14]. Dcp1a protein was not found that when equal numbers of sorted WT and -/- -/- detected in quiescent MuSC (marked by MuSC-enriched Fmr1 MuSCs were plated, there were fewer Fmr1 membrane protein Caveolin 1 (Cav) (Fig. 1L), but cells over the course of 6 days in culture compared to formed a fine striated pattern in the myofiber cytoplasm, WT (Fig. 3d). Consistent with this observation, acute largely perpendicular to expected sarcomeric knockdown of Fmrp in C2C12 myoblasts reduced clono- organization. Similarly, Xrn1 was not present in MuSC, genic performance (Fig. S3A) but did not lead to in- but exhibited a clear striated pattern in myofibers (Fig. creased cell death through apoptosis (Fig. S3B), nor were 1M). These observations reveal that components of the markers of senescence significantly induced (Fig S4). To- mRNA storage/stabilization complex (Fmrp, GW182) gether, these preliminary results indicate that Fmrp ex- are highly expressed in the nuclei of quiescent MuSC, pression is required for achieving normal fiber caliber in while the mRNA decay complex components (Dcp1a, postnatal adult skeletal muscle and that this phenotype Xrn1) are not. may be linked to a defect of knockout MuSC in prolifer- ation, reactivation from quiescence or clonogenic self- Reciprocal dynamics of Fmrp and Dcp1a puncta during renewal in culture. MuSC activation on single fibers To examine the dynamics of mRNP granules during Differential expression of mRNP proteins in quiescent, MuSC activation, we cultured isolated myofibers for up proliferating, and differentiated muscle cells in culture to 48 h and determined the pattern of expression of ,To explore the muscle cell-intrinsic functions of Fmrp, Fmrp and Dcp1a. Activation of MuSCs led to EdU in- we examined the expression of a series of mRNP pro- corporation at 24 and 48 h. Fmrp puncta were promin- teins (schematized in Fig. 4a), in a tractable adult ent in quiescent MuSC, but showed dispersal and a MuSC-derived murine C2C12 culture model that per- diffuse staining pattern after 24 h. By contrast, Dcp1a mits the generation of pure populations of proliferating puncta were not seen in quiescent MuSC, and became myoblasts (MB), quiescent myoblasts (G0), or differenti- evident in activated and proliferating MuSC (EdU+) at ated myotubes (MT) [2, 26]. In particular, this model al- 24 h (Fig. 2a–c). Thus, activation of MuSCs in their lows the entry to mitotic quiescence to be examined, a myofiber niche is associated with reciprocal changes in limitation in other similar techniques. The three cellular accumulation of mRNP granules associated with oppos- states were distinguished using expression of Myogenin ing functions, i.e., transcript turnover (Dcp1a) vs. (Myog), a master regulator of myogenic differentiation, utilization (Fmrp). and Cyclin D1, a canonical marker of proliferation: MB + - - + are Cyclin D1 Myog , MT are Cyclin D1 Myog and - - Fmrp knockout mice exhibit altered muscle stem cell G0 are Cyclin D1 Myog (Figs 4b and S5A). We investi- proliferation gated the abundance of Fmrp and Dcp1a, together with To examine whether Fmrp observed in mRNP granules other categories of mRNP components, namely: (i) pro- in quiescent MuSC is important for stem cell function teins involved in translation repression and formation of in vivo, we analyzed skeletal muscle from the Fmr1 SGs (Fmrp, Tia1) [10, 40], and translation initiation Roy et al. Skeletal Muscle (2021) 11:18 Page 10 of 28 -/- Fig. 3 ,Reduced muscle fiber caliber and MuSC proliferation in Fmr1 mice. a Cryo-sections (20 μm) of adult tibialis anterior muscle isolated from -/- wild-type (WT, left) and Fmr1 mice (right) immunolabelled with laminin (red) and nuclei counterstained with DAPI (blue): myofibers show -/- reduced diameter in Fmr1 muscle. b Left panel shows qRT-PCR quantification of mRNA encoding Fmrp isolated from whole muscle of adult WT -/- and Fmr1 mice. Values represent mean + SEM; n = 2, two-tailed paired Student’s t test is indicated. ** p < 0.01. Fmr1 RNA is detectable at lower -/- -/- levels in Fmr1 as described [30, 38, 39]. Right panel: quantification of mean myofiber cross-sectional area (CSA) in wild-type and Fmr1 muscle cryosections. Values represent mean + SD; n = 250. CSA from two mice. Two-tailed unpaired Student’s t test is indicated. *** p value < 0.0001, (N -/- + = 2 male mice per genotype). c–d. Muscle stem cells isolated from Fmr1 mice do not proliferate well in culture. c The proportion of VCAM , - -/- - + CD45 MuSC is similar in adult WT and Fmr1 mice. However, there is a noticeable reduction in the VCAM , CD45 cells suggesting effects on the -/- leukocyte compartment. d Equal numbers of FACS purified MuSC isolated from the hind limb muscle of adult WT and Fmr1 mice were plated -/- in culture for 0 or 6 days. Fmr1 cells show poor population expansion (N = 1 male mouse per genotype) (eIF-4E), (ii) proteins involved in the PB nonsense- but in MT, Fmrp was downregulated. Consistent with mediated mRNA decay pathway (Dcp1a, Pat1, and reversible suppression of translation in G0, eIF-4E, the Edc4), and (iii) proteins known to shuttle between these cap-binding component of the rate limiting translation two complexes, (Xrn1, Gw182, and Ago2) [10, 41] (Fig. initiator eIF-4F complex, is strongly downregulated in 4c, d). Briefly, the translation repressors Fmrp and Tia1 G0, but upregulated in MT. Dcp1a (and another mRNA continued to be expressed in G0 at levels similar to MB, decay factor Edc4) were less abundant in both G0 and Roy et al. Skeletal Muscle (2021) 11:18 Page 11 of 28 Fig. 4 Differential expression of mRNP granule proteins in proliferating, quiescent and differentiated muscle cells in culture. a Schematic depicts segregation of transcripts into translating and non-translating pools on emergence from the nucleus with a constellation of RNA-binding proteins. Non-translated transcripts may be sequestered in mRNPs enriched for decay complex (mRNA turnover) or storage granule components (translational repression/stabilization of mRNA). b Western blot analysis showing that three distinct cellular states can be distinguished by expression of Myogenin and Cyclin D1 (MB: asynchronously proliferating myoblasts are CycD1+, MyoG-; MT: 5-day differentiated myotubes are CycD1-, MyoG+; G0: quiescent myoblasts are CycD1-, MyoG-). c Western blot profile of mRNP granule protein expression across three cellular states. MB: proliferating myoblasts; G0: quiescent myoblasts; MT: differentiated myotubes. Expression of most proteins is suppressed in G0; notable exceptions are Fmrp and Tia1, which are maintained in G0 (see Table 1 and quantification in Fig. 4d). d Quantification of relative expression of mRNP granule proteins across the three cellular states calculated from densitometric analysis of immuno-blots. Each protein was normalized to Gapdh in the same sample, before normalizing to MB. Values represent mean + SD, n = 3, two-tailed paired Student’s t tests are indicated as [* p ≤ 0.05, ** p ≤ 0.01] Roy et al. Skeletal Muscle (2021) 11:18 Page 12 of 28 MT compared with MB. Overall, the quantitative ana- puncta re-appeared, consistent with the abundance of lysis of mRNP granule protein expression (Fig. 4c, d) re- Dcp1a protein in cycling MB (Fig. 5a). In MT, Fmrp was vealed that when comparing G0 to MB, proteins organized as small, dispersed cytoplasmic granules, while involved in mRNA turnover such as Dcp1a and Edc4, Dcp1a puncta were reduced compared to MB (Fig. 5a). were under-represented, while proteins involved in There was a general similarity in abundance of puncta in mRNA storage/stabilization/translational stalling (Fmrp, culture and in vivo: i.e., similar patterns in MT and myo- Tia1) were maintained at similar levels in G0 when com- fiber versus G0 and MuSC, suggesting an association of pared to MB. particular mRNP granule dynamics with these distinct To assess whether changes in expression of mRNP cellular states. The degree to which three additional proteins resulted from changes in expression of their mRNP proteins (Edc4, Pat1, Ago2) were organized into mRNAs, we used bioinformatic analysis of available puncta also varied between cellular states (Fig. S1). Add- transcriptome data from in vivo fixed (quiescent) satel- itional representative images (Fig. S2A) show the in- lite cells compared with activated (proliferating) satellite creased presence of Fmrp puncta in G0 and Dcp1a cells [42] and grouped them according to their function puncta in proliferation, which is supported by quantifi- [43, 44]. This comparison revealed that expression of cation of mean fluorescence intensities of these proteins genes encoding translational stalling complex proteins in subcellular puncta (Fig. S2B), and confirms the recip- (Tia1, Tncr6b, Ddx5, Ddx17) are upregulated in G0, rocal expression/organization of Fmrp and Dcp1a in G0 whereas genes encoding proteins important for mRNA compared to proliferating cells. Taken together, these decapping and turnover (Dcp1a, Edc4, Edc3) are down- immuno-localization studies indicate that translational regulated in G0 (Table 1). Fmrp expression was found to repression complexes (Fmrp, Ago2) are more prominent be maintained in quiescent satellite cells. Taken to- in G0 than in MB and MT, and that nonsense-mediated gether, this analysis suggests that compared with prolif- mRNA decay complexes (Dcp1a, Edc4, Pat1) are more erating MB, the decapping machinery is suppressed in prominent in MB and MT than in G0, both of which are G0, but that translational repression capacity is main- consistent with earlier reports of transcript stabilization tained/enhanced in G0. in quiescent cells [32, 34, 46]. Distinct organization and dynamics of mRNP granules in Global translation rates and expression of translation two mitotically inactive states initiation factors are suppressed in G0 To compare the distribution and dynamics of mRNP To compare global translation rates between the prolif- complexes in different cellular states in culture, we ex- erating, differentiated and quiescent states, we used in- amined the staining pattern of Fmrp and Dcp1a using corporation of O-propargyl-puromycin (OPP) to immunofluorescence confocal microscopy (Fig. 5a). As biosynthetically label nascent proteins (Fig. 5b, c). Rap- active mRNPs self-assemble into observable puncta and idly growing MB pulsed with OPP for 1.5 h showed disassemble upon releasing bound mRNA [36, 43], sub- strongly labeled cytoplasm and nucleoli, possibly reflect- cellular staining patterns are a reflection of the activity ing the nucleolar location of newly synthesized riboso- state of these complexes. In asynchronously proliferating mal proteins during ribosome biogenesis. In fused MT, MB, Dcp1a and Fmrp were present in small, numerous, cytoplasmic OPP labeling predominated, likely reflecting non-overlapping cytoplasmic puncta, consistent with greater synthesis of sarcomeric and other non-ribosomal their participation in distinct complexes with distinct proteins. By contrast, G0 cells showed low and variable functions. In G0, whereas Dcp1a immunolabeling was OPP labeling of the cytoplasm. Many G0 cells were es- low and diffuse (not punctate), the size and intensity of sentially unlabeled above background levels (Fig. 5b, c). cytoplasmic Fmrp granules dramatically increased, and Moreover, nucleoli could not be distinguished (Fig. 5b). nuclear-localized Fmrp granules were also prominent, These findings indicate lower rates of protein synthesis while total Fmrp protein level was maintained (Fig. 4c, and ribosome assembly in G0 cells. d), suggesting enhanced granule assembly, and greater To investigate translation by an independent method, translational repression. Notably, mRNP immuno- we analyzed expression of two translation initiation fac- detection patterns in cultured G0 cells (Fig. 5a) reflected tors, eIF-4E (the rate limiting factor in cap-dependent the patterns observed in vivo in MuSC (Fig. 1) with re- translation that also regulates mRNA export) and eIF4G spect to (i) increased Fmrp puncta and reduced Dcp1a (a scaffold for assembly of the eIF-4F complex compris- puncta and (ii) the appearance of Fmrp puncta in the G0 ing eIF-4E, eIF-4G, and eIF-4A on the 5’ cap). Both pro- nucleus. teins were present in MB and MT, but both were much ,We next tested the effect of cell-cycle reactivation on reduced in G0 (Fig. 5d), and strongly re-induced in a mRNP granules. Three hours after reactivation from qui- punctate pattern after reactivation from quiescence (R3), escence (R3) Fmrp puncta disappeared and Dcp1a when protein synthesis begins to recover (Fig. 5 c, d). Roy et al. Skeletal Muscle (2021) 11:18 Page 13 of 28 Table 1 Bio-informatic analysis of transcripts encoding mRNP components Data from Yue et al (2020) Accession GSE113631 mRNP component genes FPKM values fSC vs ASC Table 1 Functional Class Gene ASC_ ASC_ fSC_ fSC_ Fold change adjusted p-value (BH Fold Symbol rep1 rep2 rep1 rep2 (Log2) correction) change ARE binding Tnpo1 57.211 65.710 17.063 15.135 -1.93 0.0353 0.26 ARE binding Zfp36 22.033 19.558 15.694 13.194 -0.53 0.1265 0.69 ARE binding Tia1 7.034 6.693 16.816 16.464 1.28 0.0117 2.42 Deadenylation Cnot1 105.514 110.661 20.399 21.317 -2.37 0.0135 0.19 Deadenylation Cnot2 41.603 42.019 26.262 15.837 -0.99 0.1112 0.50 Deadenylation Cnot6l 32.610 34.390 24.860 25.916 -0.4 0.0481 0.76 Deadenylation Cnot8 25.708 26.029 30.524 23.093 0.05 0.8439 1.04 Deadenylation Pan2 12.528 10.940 15.276 14.178 0.33 0.1543 1.26 Deadenylation Tob2 11.102 12.033 16.704 14.161 0.42 0.1717 1.34 Deadenylation Cnot4 10.540 8.787 11.628 17.230 0.58 0.3379 1.49 Deadenylation Pan3 3.337 3.622 5.453 5.373 0.64 0.0281 1.56 Deadenylation Cnot3 0.864 0.889 3.501 3.168 1.93 0.0252 3.81 Decapping Lsm3 249.390 284.670 85.416 78.982 -1.7 0.0357 0.31 Decapping Dcp1a 16.651 17.486 5.447 5.466 -1.65 0.0151 0.32 Decapping Edc3 12.841 12.809 5.918 4.803 -1.26 0.0274 0.42 Decapping Patl1 10.775 9.751 5.821 4.656 -0.97 0.0601 0.51 Decapping Edc4 24.210 23.716 12.896 12.523 -0.91 0.0130 0.53 Decapping Dcp2 9.174 10.101 7.586 8.180 -0.29 0.1489 0.82 Decapping Lsm2 32.104 34.063 28.098 31.387 -0.15 0.3121 0.90 Decapping Lsm6 209.767 214.354 204.279 179.065 -0.15 0.3451 0.90 Decapping Lsm7 58.943 65.467 66.429 58.668 0.01 0.9582 1.01 Decapping Lsm5 87.274 87.571 126.064 111.808 0.44 0.0971 1.36 Decapping Lsm1 4.656 4.518 7.664 5.746 0.55 0.2374 1.46 Decapping Dcp1b 0.440 0.394 2.564 2.611 2.63 0.0081 6.20 Decapping Lsm4 5.825 5.767 42.218 38.352 2.80 0.0207 6.95 Many RNP functions Eif4e 230.579 234.878 57.512 38.165 -2.28 0.0202 0.21 Many RNP functions Ddx1 56.948 56.634 16.101 9.526 -2.15 0.0276 0.23 Many RNP functions Eif2s2 183.202 191.179 68.677 56.697 -1.58 0.0214 0.33 Many RNP functions Eif2b1 76.122 77.151 33.579 23.721 -1.42 0.0379 0.37 Many RNP functions Gemin5 70.000 66.713 28.690 30.406 -1.21 0.0184 0.43 Many RNP functions Trim59 4.600 4.463 2.249 1.956 -1.11 0.0247 0.46 Many RNP functions Lima1 15.817 12.996 8.282 10.047 -0.65 0.1513 0.64 Many RNP functions Dhx40 5.881 6.893 5.668 4.463 -0.33 0.3256 0.79 Many RNP functions Pdlim7 15.903 17.440 19.606 18.779 0.20 0.1690 1.15 Many RNP functions Fmr1 42.281 42.132 58.735 49.227 0.35 0.2056 1.28 Many RNP functions Xrn1 11.279 11.434 14.211 17.589 0.49 0.1853 1.40 Many RNP functions Eif2b4 22.085 21.683 33.687 31.827 0.58 0.0322 1.50 Many RNP functions Pcbp2 77.519 73.659 138.630 159.083 0.98 0.0542 1.97 Many RNP functions Trim28 19.447 19.682 33.020 45.418 1.00 0.1500 2.00 Many RNP functions Lpp 30.789 32.101 113.147 97.692 1.75 0.0385 3.35 Many RNP functions Ddx5 65.868 58.928 263.633 201.650 1.90 0.0746 3.73 Roy et al. Skeletal Muscle (2021) 11:18 Page 14 of 28 Table 1 Bio-informatic analysis of transcripts encoding mRNP components (Continued) Data from Yue et al (2020) Accession GSE113631 mRNP component genes FPKM values fSC vs ASC Table 1 Functional Class Gene ASC_ ASC_ fSC_ fSC_ Fold change adjusted p-value (BH Fold Symbol rep1 rep2 rep1 rep2 (Log2) correction) change Many RNP functions Peg3 105.426 104.905 445.266 466.449 2.12 0.0135 4.33 Many RNP functions Ybx1 64.225 64.993 464.325 296.992 2.56 0.1191 5.90 Many RNP functions Ddx17 9.400 8.959 111.803 114.915 3.63 0.0081 12.35 Many RNP functions Trim21 0.015 0.017 0.417 0.613 5 0.0809 32.00 miRNA-mediated gene Limd1 46.125 42.266 43.565 43.929 -0.01 0.8565 0.99 silencing miRNA-mediated gene Tnrc6a 8.243 8.496 10.950 13.634 0.55 0.1674 1.47 silencing miRNA-mediated gene Htt 2.976 2.676 6.126 6.307 1.14 0.0195 2.20 silencing miRNA-mediated gene Tnrc6b 4.196 4.958 14.202 15.669 1.71 0.0293 3.27 silencing miRNA-mediated gene Ipo8 4.524 4.272 16.861 15.367 1.87 0.0242 3.66 silencing miRNA-mediated gene Tnrc6c 3.082 3.039 16.137 17.995 2.48 0.0247 5.58 silencing NMD pathway Smg7 58.640 66.881 11.867 12.560 -2.36 0.0299 0.19 NMD pathway Upf2 13.025 15.822 16.257 16.691 0.19 0.3801 1.14 NMD pathway Upf1 14.745 16.093 17.029 20.119 0.27 0.2885 1.21 NMD pathway Pnrc2 44.583 42.606 61.898 49.546 0.35 0.2770 1.28 NMD pathway Upf3b 24.102 22.611 42.912 44.988 0.91 0.0230 1.88 NMD pathway Smg5 2.169 2.287 7.059 6.539 1.61 0.0217 3.05 NMD pathway Smg6 6.228 6.373 26.228 27.657 2.10 0.0148 4.28 NMD pathway Upf3a 5.788 5.261 34.781 29.334 2.54 0.0379 5.82 NMD pathway Pnrc1 3.468 2.819 103.360 88.526 4.93 0.0293 30.52 The altered abundance of eIF-4E was consistent with the maintenance of quiescence [48], the data above sug- our western analysis (Fig. 4 c, d). These initiation fac- gest that a number of G0-induced transcripts may also tors showed some nuclear localization in G0, which be sequestered in non-polysomal compartments, to be was greatly enhanced during synchronous reactivation, mobilized for protein synthesis required for the return but not detected in either cycling MB or MT, pos- to the cell cycle [17]. To visualize ongoing translation sibly reflecting involvement in upstream functions activity directly, we analyzed steady-state polysome pro- such as mRNA export, that are important for cell files in each cellular state. To ensure polysome integrity cycle re-entry (Fig. 5d). Taken together, these findings during isolation and display, cells were treated briefly are consistent with the notion of G0 as a state where with cycloheximide (CHX) prior to lysis, to arrest trans- global translational repression is coupled to mRNA lating ribosomes on mRNAs, followed by separation on stabilization in granules, keeping cells primed for cell sucrose density gradients (Fig. 6a–c). The profile of cycle re-entry [17, 34]. RNA-protein complexes was quantified in density- separated fractions and analyzed by immuno-blotting. A Quiescent myoblasts exhibit puromycin-resistant mRNP second profile was run from cells in each state that were complexes in G0 treated briefly with puromycin (Puro), that successfully ,Quiescent cells show low transcriptional activity com- disengaged mRNA from translating ribosomes, removing pared with proliferative or differentiated states. Although the polysome profile (Fig. 6a–c). With respect to mRNP many transcripts are specifically induced in G0 [1, 47] granule dynamics, CHX rapidly dissociates mRNP gran- and some must be translated into proteins required for ules, whereas Puro promotes their assembly [10]. Roy et al. Skeletal Muscle (2021) 11:18 Page 15 of 28 Fig. 5 Assembly of mRNP into puncta in different cellular states correlates with levels of protein synthesis. a Representative immunofluorescence images showing Fmrp (green) and Dcp1a (red) puncta in G0, MB, and MT, as well as cells reactivated for 3 h from G0 (R3). Arrows indicate prominent puncta. Notably, Fmrp puncta are large and prominent in G0, disperse at 3 h post reactivation and are less evident in asynchronous MB. Dcp1a puncta are nearly absent in G0 and reappear at R3; Dcp1 puncta are also more prominent in MB than MT. Fmrp and Dcp1a mostly localize to distinct puncta; the rare yellow puncta seen in R3 and MT may reflect transient co-localization due to passage of transcripts between two types of mRNP granules as inferred in [45]. b Measurement of the rate of protein synthesis using OPP incorporation into newly synthesized proteins reveals active translation in MB and MT, and substantial suppression in G0. c Quantification of images in b. Fluorescent intensity was measured in 150 cells from each condition, the box and whisker plot shows integrated fluorescence for each cell (each dot represents one cell), limits on the box correspond to 75th and 25th percentile values. “Mb-neg” and “G0-neg” represent samples of MB and G0 that were not pulsed with OPP but processed for detection along with samples that were exposed to OPP. N = 2 biological replicates. Data were analyzed by multivariate ANOVA with post hoc HSD Tukey tests performed for each pairwise comparison. *** p < 0.0001, ** p < 0.001. d Immunolabeling of translation initiation factors eIF-4E (red) and eIF-4G (pink) in G0, R3, MB, and MT: Upper panel shows merged images, and lower panels show detection of each factor individually. Expression and assembly of these translation factors correlates with levels of protein synthesis seen in b and c: poor in G0, restored assembly with distinct puncta in R3, and strong expression and organization of eIF-4E and eIF-4G complexes in MB and MT Roy et al. Skeletal Muscle (2021) 11:18 Page 16 of 28 Fig. 6 Polysome profiles of proliferating, quiescent, and differentiated muscle cells reveal stalled polysomes in G0. Translational profiles of myoblasts (a), myotubes (b), and G0 cells (c) using polysome display on sucrose gradients. Panels on left depict profiles derived from cells briefly treated with CHX to ‘fix’ ribosomes in the act of translation, while panels on the right depict profiles derived from cells treated with Puro to disrupt translation by mRNA release. Western blotting of proteins isolated from 9 individual 1-ml fractions from the sucrose gradients (equal volumes loaded) reveals (i) distribution of ribosomes in each fraction based on ribosomal protein P0 (middle) and the extent of association of decay complex based on Xrn1 (top), and translation inhibitory complex based on Fmrp (botttom) with each fraction. Comparison of the profiles and distribution of individual proteins reveals very poor translation in G0, correlating with the OPP incorporation in Fig. 5. The presence of puromycin-insensitive complexes in G0 arrested cells, suggests polysome stalling. d Analysis of transcript distribution in polysome profiles correlates with rate of protein synthesis and suggests low mRNA utilization in G0. qRT-PCR analysis of selected transcripts (GAPDH, Cyclin D1, MyoD, Myf5, and p27) from RNA isolated from the mRNP-, monosome-, and polysome-containing fractions of profiles depicted in Fig. 6a–c. All transcripts tested show substantial enrichment in the mRNP and monosome compartment in G0 compared with the monosome and polysome fraction, suggesting a severe suppression of protein synthesis consistent with the OPP incorporation study (Fig. 5b, c). Notably, none of the transcripts tested show appreciable enrichment in the mRNP fraction in MB and MT, indicating their robust translational utilization in the polysomal compartment. Values represent the mean + SD of transcript levels in fractions from two independent polysome profiles for each condition Roy et al. Skeletal Muscle (2021) 11:18 Page 17 of 28 In MB, in addition to the strong ribosomal subunit Table 2 Non-translating to translating ratio (mRNP: mono + poly) ratio in 3 different states (MB, MT, and G0) peaks in fractions 3, 4, 5 (containing free 40S, 60S sub- units, and 80S monosomes, respectively), a range of State Transcript polysome peaks was visible (fractions 6–9), which was Cyclin D1 p27 MyoD Myf5 Gapdh sensitive to Puro treatment, showing that the cells were MB 0:100 0:100 0:100 2:98 3:97 engaged in active translation (Fig. 6a). Western blot pro- MT 0:100 0:100 0:100 0:100 0:100 files confirmed that the ribosome-containing heavier G0 70:30 60:40 50:50 65:35 50:50 fractions (3–9, marked by the presence of the ribosomal protein P0) were largely devoid of mRNP proteins Xrn1 and Fmrp, which were enriched in the non-polysomal Cdkn1b/p27), as seen in global transcript analysis de- fractions 1–2. On treatment with Puro, disruption of rived from transcriptome data (Venugopal et al, 2020 polysomes was evident and accompanied by loss of P0 [49]) (Fig. S5B and Table 2). Consistent with the bulk protein from fractions 6–9 (Fig. 6a). MT also displayed polysome profile that shows low polysome assembly in very active translation, showing monosome and poly- G0, all transcripts tested show substantial enrichment in some peaks similar to the profile in MB (Fig. 6b). Simi- mRNP and monosome compartments and < 10% in the larly, Xrn1 and Fmrp were detected at low levels in polysome fraction in G0 cells (Fig. 6d). By contrast, in fraction 6, but otherwise, these mRNP proteins were both MB and MT, all five mRNAs were enriched on largely absent from the polysome fractions 7–9. As in polysomes, with barely detectable presence in the mRNP proliferating MB, Puro treatment led to the loss of poly- fraction, with Fig S5B reflecting the global transcript sta- some peaks in MT. tus of the mRNAs relative to MB consistent with the In G0 cells, by contrast, polysomes were nearly un- high rates of protein synthesis typical of these states. detectable and fewer monosomes were seen, consistent Thus, the observation that, for all five transcripts tested, with the accumulation of P0 in the lighter complexes the majority (as a proportion of each transcript’s abun- (Fractions 1–4 (Fig. 6c). Nevertheless, in G0 cells, P0 dance in each state), was either found in large polysomes persisted in high molecular weight complexes (Fractions in MB (i.e., was actively translated), shifted to mono- 6–9 in Puro vs CHX), that were insensitive to Puro (Fig. somes compared with polysomes in MT (i.e., was less 6c). Together, these observations suggest the presence of actively translated), or further shifted to mRNPs in cells heavy mRNP complexes in G0 cells that are not engaged in G0 (i.e., was not translated at all), we interpret to re- in active translation. These heavy mRNPs could be flect the global translational status of each state. Taken stalled polysomes, or mRNA captured in other heteroge- together with the repressed global rates of protein syn- neous paused complexes along with ribosomes, but not thesis and increased accumulation of mRNP proteins in undergoing active translation. Treatment of G0 cells visible puncta, we conclude that mRNA sequestration in with Puro increased mRNPs in the heavy fractions 7–9, a non-translated compartment is a broad regulatory the opposite of the effect of Puro in MB (Fig. 6a, c). The process that is enhanced in reversible G0, but not in sustained enrichment in the heavier fractions (7–9) in post-mitotic MT. puromycin-treated G0 cells suggests that ribosomal pro- teins are present in non-canonical high molecular weight Dcp1a and Fmrp reciprocally regulate their protein complexes in G0 cells, which are absent in MB and MT. abundance and granule assembly Taken together with OPP incorporation and eIF-4E ex- As Fmrp and Dcp1a are known to regulate distinct as- pression levels, these results demonstrate that proliferat- pects of mRNA function (translation vs. turnover [15]) ing and differentiated cells are actively engaged in and were found in different complexes, we considered translation, while quiescent cells show markedly sup- the possibility that these proteins might also cross- pressed protein synthesis, potentially associated with se- regulate. We used siRNA-mediated knockdown to per- questered and stalled ribosomes. turb the levels of each protein and evaluated the effect of knockdown of one protein on abundance of the other Transcripts accumulate in a non-polysomal mRNP protein using western blotting (Fig. 7a). Proliferating compartment specifically in G0 myoblasts were transfected with siRNA smart pools To probe the distribution of specific transcripts between (comprising four independent siRNAs) designed to tar- actively translating and inactive sequestered compart- get either Dcp1a or Fmr1 mRNAs. A non-targeting ments, we used qRT-PCR analysis on RNA isolated from siRNA pool was used as a control. Knockdown efficiency the mRNP, monosome- and polysome-containing frac- was confirmed to be 70–85% for Fmrp and 40–50% for tions (Fig. 6d). We selected mRNAs whose levels are (i) Dcp1a by western blotting (Fig. 7a). Indeed, knockdown unchanged (Gapdh) (ii) suppressed in G0 (Cyclin D1, of Fmrp led to an induction of Dcp1a protein levels and MyoD), or (iii) maintained/induced in G0 (Myf5, vice versa, knockdown of Dcp1a was accompanied by Roy et al. Skeletal Muscle (2021) 11:18 Page 18 of 28 Fig. 7 (See legend on next page.) higher levels of Fmrp (Fig. 7a). This reciprocal regulation knockdown of Fmrp was readily observed as reduced im- at the level of protein abundance was accompanied by munofluorescence, and accompanied by an enhanced in- increased detection of the respective protein in cytoplas- tensity of Dcp1a, and reciprocally, knockdown of Dcp1a, mic puncta (Fig. 7b). Quantification of the fluorescent led to loss of Dcp1a detection and enhanced intensity of intensity of cytoplasmic staining (Fig. 7c) revealed that Fmrp. Taken together, these experiments reveal cross- Roy et al. Skeletal Muscle (2021) 11:18 Page 19 of 28 (See figure on previous page.) Fig. 7 Cross-regulation of Fmrp and Dcp1a in knockdown myoblasts. a Following transfection with either Dcp1a or Fmr1 siRNA pools or the control pool (Scr), knockdown myoblasts were incubated in growth medium for 18 h. Left: Immunoblot analysis shows that in Fmrp knockdown, Fmrp abundance is reduced but Dcp1a expression is enhanced. Likewise, Fmrp protein levels are increased in Dcp1a knockdown. Values depicted under each lane represent protein levels from normalized densitometric scans, relative to level in Scr. Right panel: Densitometry of western blots of Dcp1a, and Fmrp proteins normalized relative to Scr with Gapdh as internal control. Bar graphs represent mean ± SD from n = 3. Two-tailed paired Student’s t tests are indicated as * p <0.05. ** p <0.01. b Knockdown effects on Dcp1a and Fmrp by their respective targeting siRNAs are detectable at the subcellular level. Consistent with changes at the level of protein abundance, immunofluorescence analysis shows that increased Dcp1a expression in Fmr1 knockdown is accompanied by enhanced Dcp1a puncta assembly, while compromising Dcp1a expression leads to enhanced Fmrp puncta assembly. Scale bars represent 15 μm except in zoomed panels where scale bars represent 8 μm. c Quantitative image analysis of b:The fluorescence intensities of Fmrp and Dcp1a following immunostaining of Scr, siFmr1, and siDcp1a samples were calculated and represented as box and whisker plots. Significant increases in Dcp1a puncta were observed in Fmrp knockdown and vice versa increased Fmrp puncta were observed in Dcp1a knockdown. Data were obtained from triplicate samples for each condition and graph shows scoring of at least n = 75 cells each for Scr (Scrambled), siFmr1 (Fmrp knockdown), siDcp1a (Dcp1a knockdown). Two-tailed paired Student’s t test results are indicated as *** p <0.001 regulation of Fmrp and Dcp1a not only at the level of Knockdown of Fmrp and Dcp1a on cell cycle regulators protein abundance, but also at the level of protein as- during quiescence and reactivation sembly into puncta. As knockdown of Fmrp and Dcp1a had opposing effects on MB proliferation, we evaluated the consequences of the knockdowns on expression of cell cycle regulators in Fmrp and Dcp1a play opposing roles in control of MB G0 and R3. In G0 conditions, we found nearly 5-fold in- proliferation crease in Cyclin A2 protein expression when Dcp1a was The results so far indicate differential mRNP granule pro- knocked down (Figs. 8e and S6C), consistent with the in- tein abundance and assembly in distinct cellular states: the creased proliferation observed in the MB condition. quiescent state is enriched in translational silencing/re- Fmrp knockdown, however did not affect protein ex- pressive complexes, whereas proliferating cells are pression of Cyclin A2 or Cyclin E (Figs. 8e and S6C), enriched in the classical mRNA decapping and decay consistent with the unchanged EdU incorporation in ei- complexes. Further, reducing the abundance of transla- ther knockdown in G0 (Fig. 8c). Moreover, Fmrp knock- tional repressor Fmrp by knockdown led to increased down G0 cells displayed reduced levels of mRNAs abundance and assembly of mRNA decay regulator Dcp1a encoding Cyclin A2, B1, E, D1, and ki67, but negligible and vice versa. To determine whether the differential en- change in the levels of transcripts encoding either cell richment of the decay and repressive complexes plays a cycle inhibitors (Cdkn1a/p21, Cdkn1b/p27) or myogenic role in the maintenance of a particular cellular state, we regulatory factors (MyoD, Pax7 and Myf5) (Fig. S7). By examined the phenotypes of the knockdown cells. Knock- contrast, Dcp1a knockdown increased levels of pro- down of Dcp1a in proliferating MB caused cells to prolif- proliferative transcripts including ki67 and Cyclin A2, erate more rapidly than control siRNA-treated cells, as B1, E, and ki67, and reduced levels of the anti- evidenced by a significant increase in cell number and 5- proliferative Cdk inhibitor, Cdkn1a/p21, consistent with Ethynyl-2´-deoxyuridine (EdU) incorporation (Fig. 8a). By increased proliferation. Interestingly, both MyoD and contrast, knockdown of Fmrp led to reduced EdU incorp- Pax7 transcript levels were strongly reduced. Taken to- oration (Fig. 8a), mimicking the reduced proliferative gether, the reciprocal molecular phenotypes of Fmrp -/- capacity seen in MuSC from Fmr1 mice (Fig. 3d). To- and Dcp1a knockdowns in cells in G0 were consistent gether, these results indicate that compromising the ex- with the observed reciprocal effect on proliferation. pression of key decapping and repressive/silencing mRNP Despite the altered mRNA profiles detailed above, proteins differentially affects proliferation in myoblasts. both Dcp1a and Fmrp knockdown cells were able to Dcp1a and Fmrp thus exert opposing effects on cell prolif- undergo arrest as evidenced by the absence of DNA syn- eration, possibly by targeting different transcripts for deg- thesis in suspension culture. However, whereas Dcp1a radation, translational repression, and/or sequestration. knockdown cells re-entered the cell cycle more rapidly To identify the regulatory nodes at which Fmrp and (see below), Fmrp knockdown cells were deficient in re- Dcp1a might exert their effects, we evaluated the expres- activation. These results suggest that Fmrp may nor- sion of cell cycle regulatory proteins by western blotting. mally enable reversible quiescence. Therefore, we In Dcp1a knockdown cells, Cyclin A2 protein expression determined whether Fmrp knockdown cells activated increased, consistent with enhanced proliferation (Figs. 8e other inhibitory programs, such as apoptosis or senes- and S6A). In Fmrp knockdown cells, by contrast, Cyclin cence. Flow cytometry of Fmrp knockdown cells re- A2 and Cyclin E protein levels were decreased, consistent vealed no increase in markers of apoptosis (Fig. S3B). with reduced proliferative capacity (Figs. 8e and S6A). Further, there was no increase in markers of senescence Roy et al. Skeletal Muscle (2021) 11:18 Page 20 of 28 Fig. 8 Knockdown of Fmrp and Dcp1a show opposing effects on the cell cycle. Proliferating myoblasts (MB) were treated with siRNAs (Scr, siDcp1a, siFmr1) for 18 h and either induced to enter G0 for 48 h or induced to differentiate for 48 h. For reactivation, G0 cells were harvested and plated on dishes or coverslips for 3 h. a–d EdU incorporation in MB (a), MT (b), G0 (c), and R3 (d): Dcp1a knockdown cells show increased incorporation of EdU in MB, MT, and R3, but not in G0, while Fmrp knockdown cells show decreased incorporation. EdU assay was performed simultaneously for all the conditions. Graphs show quantification by scoring > 500 nuclei for each condition in 3 biological replicates. * p < 0.05, ** p < 0.01. Two-tailed paired Student’s t test was performed. e Consistent with EdU incorporation, Cyclin A2 and Cyclin E show altered protein expression in Dcp1a and Fmrp knockdowns. Gapdh used as internal control. Values represent the mean + SD in 3 biological replicates (Fig. S4). Yet, as referred to earlier, Fmrp knockdown Indeed, knockdown of Dcp1a and Fmrp had a cells showed poor colony-forming ability, indicating marked impact during reactivation of quiescent cells. compromised self-renewal (Fig. S3A). Taken together, At 3 h of reactivation, Dcp1a knockdown cells preco- these results suggest that the absence of Fmrp leads to ciously displayed increased EdU incorporation com- an altered G0 state that we term “aberrant quiescence”. pared with control cells (Fig. 8d). Supporting the Roy et al. Skeletal Muscle (2021) 11:18 Page 21 of 28 Fig. 9 Knockdown of Fmrp and Dcp1a show similar effects on differentiation. Proliferating myoblasts (MB) were treated with siRNAs (Scr, siDcp1a, siFmr1) for 18 h and induced to differentiate for 48 h. a Both Fmr1 and Dcp1a knockdowns show reduced number of Myogenin nuclei. Upper Panel: Immunofluorescence of Myogenin (MyoG) and Fmrp in Scr, siFmr1 and siDcp1a. Scale bars represent 35 μm except in magnified panels where scale bars represent 17 μm. Lower panel: quantification based on 3 replicates, with > 600 nuclei scored per condition. b Knockdown of either Fmrp or Dcp1a affects fusion of myoblasts as shown by reduction in fusion index. Upper panel: immunofluorescence of myosin heavy chain (Myosin HC) and Fmrp in Scr, siFmr1, and siDcp1a. Lower panel: fusion index calculated as the ratio of the number of nuclei in myotubes with 2 or more nuclei over the total number of nuclei × 100 for n = 3 biological replicates. More than 850 nuclei were counted per condition. For a and b * p < 0.05, ** p < 0.01. Two-tailed Student’s t test was performed. c Representative western blots (from 3 biological replicates) of Myogenin (MyoG) and Myosin Heavy Chain (Myosin HC) proteins in MB and MT; Gapdh is internal control. d Densitometry of western blots of Myogenin (MyoG) and Myosin Heavy Chain (Myosin HC) proteins in MB and MT in c; Gapdh is loading control. Western blot analysis decreased expression of both myogenin and myosin when either Fmrp or Dcp1a expression is reduced. Two-tailed Student’s t test was performed, ** p < 0.01, *** p < 0.001. Values represent the mean + SD in 3 biological replicates premature entry into S phase, Dcp1a knockdown cells and displayed strong suppression of Cyclin E protein showed increased expression of both Cyclin A2 and levels (Figs. 8e and S6D), indicative of poor reactiva- Cyclin E proteins (Figs. 8eand S6D).Bycontrast, tion. Together, these data are consistent with oppos- Fmrp knockdown cells showed even less EdU cells ing effects of Fmrp and Dcp1 on proliferation, and than the control at this early activation time point, suggest that Fmrp and Dcp1a modulate quiescence Roy et al. Skeletal Muscle (2021) 11:18 Page 22 of 28 Fig. 10 Model showing how cross-regulation of Dcp1a and Fmrp alters the balance of mRNA turnover and translation. The balance of Dcp1a and Fmrp are hypothesized to control the turnover and translation of different sets of transcripts in distinct cellular states (middle row). When wild- type proliferating myoblasts (MB, center) enter quiescence (G0, left), protein synthesis is repression and stalled polysomes are detected, paralleled by enrichment of the translational repressor Fmrp into prominent puncta, whereas Dcp1a puncta diminish. In contrast, differentiation (MT, right) is associated with a reduction of both Fmrp and Dcp1a puncta, suggesting a new set point for the balance of these regulators. Perturbing expression of Dcp1a (upper row) or Fmrp (lower row) has reciprocal effects on mRNP granules, and opposing phenotypic consequences. Depletion of Dcp1a leads to increased Fmrp accumulation and assembly, whereas depletion of Fmrp leads to increased Dcp1a accumulation and assembly. Dcp1a knockdown (upper row) may increase levels of proteins that enhance cell proliferation directly (via reduced mRNA turnover), and indirectly act via increasing Fmrp to reduce translation of negative cell cycle regulators. Such hyper-proliferative Dcp1a knockdown cells are resistant to induction of quiescence. Conversely, Fmrp knockdown (lower row) may increase levels of proteins that repress the cell cycle directly (via de-repressed translation), and indirectly decrease levels of proteins that positively regulate the cell cycle (via increased Dcp1a and increased turnover of transcripts). Thus, Fmrp knockdown cells show reduced cell proliferation and Dcp1a knockdown cells show increased proliferation. The observations that normal induction of quiescence leads to increased Fmrp accumulation, whereas forced suppression of Fmrp also decreases proliferation, suggest that a threshold of Fmrp accumulation/assembly is required to balance between proliferation and quiescence. The observation that depletion of either Dcp1a or Fmrp leads to compromised differentiation may be explained by altered net translation of different sets of pro- and anti-myogenic target transcripts entry/exit potentially by targeting stability/utilization correlated with a pronounced suppression of Myogenin of cyclin transcripts. protein in the same sample, consistent with translation suppressive function of Fmrp. However, maintenance of Knockdown of either Fmrp or Dcp1a compromises the knockdown cells in differentiation conditions myogenic differentiation showed that loss of either Fmrp and Dcp1a negatively af- To assess the effects of depletion of Fmrp and Dcp1a on fected differentiation as evidenced by reduced Myogenin myogenesis, knockdown myoblasts were induced to dif- protein abundance, decreased frequency of Myogenin ferentiate for 2 days. As in proliferative conditions, nuclei, reduced Myosin Heavy Chain protein expression Dcp1a knockdown in low serum conditions also led to and significantly reduced fusion index (Figs. 9a–d and sustained EdU incorporation with a corresponding in- S6E). Taken together, these results indicate that despite crease in Cyclin A2 protein (Fig, 8b, e and S6B). By con- their opposing effects on the cell cycle, optimal levels of trast, Fmrp knockdown lead to negligible EdU both Dcp1a and Fmrp are required for myogenesis. incorporation accompanied by drastic reduction in Cyc- In summary, our data support a model (Fig. 10), where lin A2 protein compared with control. Notably, the Fmrp and Dcp1a reciprocally regulate each other at the cross-regulation of Fmrp by Dcp1a knockdown (see pre- level of protein abundance and granule assembly, differ- vious section) was most pronounced in myotubes, and entially regulate the expression of cell cycle and Roy et al. Skeletal Muscle (2021) 11:18 Page 23 of 28 myogenic proteins and thereby play critical and oppos- cells enter a suppressed state that is poised for reactiva- ing roles in the transitions between proliferation and re- tion. Strikingly, proteins involved in mRNA degradation versible quiescence, ultimately leading to compromised are enriched in MB, while G0 cells are enriched in pro- differentiation. teins involved in mRNA storage/suppression of transla- tion. In particular, quiescence is characterized by Discussion reduced expression of initiation factors, low rates of pro- In this study, we show that components of mRNP gran- tein synthesis, and potentially, stalled polysomes. ules regulate MuSC proliferation and differentiation Assembly of mRNP components into granules also dif- in vitro and myogenesis in vivo, likely through changes fers between G0 and MT. mRNP granules are assembled in the translation and turnover of mRNAs encoding key around distinct transcripts and modulate their function- regulators of MuSC dynamics. ality. These mRNP-associated transcripts may either be degraded, or remain in a stable, untranslated state, Quiescent cells display distinct mRNP complexes where the composition of a particular mRNP complex Non-dividing cells are well-known to exhibit reduced determines the fate of individual transcripts. Our study macromolecular metabolism. Here, we show that muscle reveals that in culture, mRNP granules containing dec- cells in two distinct states of cell cycle arrest elaborate apping proteins of the classical decay complex (Dcp1a, distinct mRNP granule protein expression, correlating Pat1, Edc4) (Table 3) are enriched in MB, suggesting with global protein synthesis. When MB enter perman- that ‘stockpiling’ of inactive transcripts in quiescence as ent arrest associated with differentiation to MT, robust during embryogenesis [50], may facilitate cell cycle reen- levels of protein synthesis sustain tissue-specific func- try when translation resumes. Notably, during G0, trans- tions. However, in reversible arrest (G0), which is typical lationally repressive complexes (Fmrp+) dominate, of adult stem cells, protein synthesis is restricted, and consistent with the enrichment of Fmrp+ storage Table 3 Details of antibodies used in this study Antibody Species Company Catalog # Dilution for WB Dilution for IFA Argonaute2 Rabbit CST C34C6 1 in 1000 1 in 200 Cyclin A2 Rabbit Abcam ab181591 1 in 2000 Cyclin E Rabbit Abcam ab71535 1 in 1000 Dcp1a Mouse Santa Cruz sc100706 1 in 200 1 in 50 Edc3 Mouse Santa Cruz sc365024 1 in 200 1 in 50 Edc4 Rabbit Santa Cruz ab72408 1 in 200 1 in 50 eIF4E Mouse Santa Cruz sc271480 1 in 200 1 in 50 Eif4g Goat Santa Cruz sc9602 1 in 200 1 in 50 Fmrp Rabbit Sigma 4055 1 in 1000 1 in 200 FxR1 Goat Abcam ab51970 1 in 1000 Gapdh Mouse Abcam ab8245 1 in 10000 GFP Chicken Abcam ab13970 1 in 300 GW182 Mouse Santa Cruz sc56314 1 in 200 1 in 50 γH2AX (Ser139) Mouse Upstate 05636 1 in 200 MyoD Mouse Dako M3512 1 in 250 1 in 200 MyoG Mouse Santa Cruz sc12732 1 in 250 1 in 250 Myosin Heavy Chain Mouse Hybridoma A4.1025 1 in 1 1 in 1 Pabp1 Rabbit Abcam ab21060 1 in 50 Pax7 Mouse Aviva ARP30947_P050 1in 100 RLP0 Rabbit Abcam ab101279 1 in 1000 Tia1 Mouse Santa Cruz sc166247 1 in 200 1 in 50 TiaR Goat Santa Cruz sc1749 1 in 200 1 in 50 Xrn1 Rabbit Sigma sab4200028 1 in 1000 1 in 200 Roy et al. Skeletal Muscle (2021) 11:18 Page 24 of 28 granules in quiescent muscle stem cells in vivo [17] (Fig. Dcp1a protein abundance and increased formation of 1). Our findings confirm the recent report [23] that Dcp1a puncta in Fmrp knockdown cells. Fmrp is required for MuSC function in vivo. By contrast, Dcp1a knockdown enhanced S phase mRNP puncta are thought to represent sites where the entry, and enhanced mRNA levels of positive regulators mRNP granule proteins exert their function [41, 51]. of progression – Cyclins A2, B2, and D1. These changes The increased abundance of Fmrp puncta in G0 may may be a direct effect of reduced cyclin mRNA turnover. suggest a role either in the entry into or maintenance of However, given the concomitant increase in Fmrp ex- quiescence. By contrast, the reduced Dcp1a puncta pression and puncta, indirect effects of Dcp1a knock- would suggest that Dcp1a either opposes or is not im- down on negative regulators of the cell cycle cannot be portant for quiescence. As discussed in detail below, the ruled out. For example, reduced translation of a poten- functional data is in apparent contradiction with this in- tial Fmrp target such as Cdkn1a/p21 would synergize terpretation: knocking down Fmrp expression (leading with increased cyclin mRNA expression to enhance S to lower Fmrp puncta accumulation) slows the cell cycle, phase entry. The identity of direct and indirect targets of whereas knocking down Dcp1a hastens the cell cycle. Fmrp and Dcp1a in different cellular states are currently When Fmrp expression is compromised, the cells enter not known, and would likely resolve this conundrum. into an aberrant quiescence, from which they are unable The opposing phenotypes of the Dcp1a and Fmrp to exit. The aberrant quiescent state is not accompanied knockdown in MB are consistent with the opposing by increased cell death or activation of senescence roles played by these proteins in regulation of the cell markers, but does exhibit compromised clonogenicity cycle. These phenotypic changes are sustained in both (self-renewal), and warrants further investigation. A pos- quiescent as well as reactivated conditions. Specifically, sible explanation is that Fmrp plays a role in the transla- increased pro-proliferative transcripts and decreased cell tional pausing we observed in the primed or poised cycle inhibitor transcripts are observed in Dcp1a knock- quiescent state, as first reported by Crist et.al., [17]. If in down cells in G0, and the converse in Fmrp knockdown absence of Fmrp its target mRNAs are continuously cells. However, the increased Cyclin A2 protein expres- translated, the cell might be unable to leave quiescence. sion in Dcp1a knockdown is not accompanied by in- Another possibility is that, as Dcp1a expression and as- creased EdU incorporation in G0, suggesting that other sembly are enhanced in Fmrp knockdown cells, tran- elements act to maintain quiescence. scripts that would normally be stabilized in a The effects of Fmrp knockdown on the activation out translationally repressed state (associated with Fmrp) of quiescence in culture were reminiscent of the re- -/- now become targets for more rapid turnover by Dcp1a. stricted proliferation of Fmr1 MuSCs, reflected by un- We hypothesize that among these destabilized tran- changed EdU incorporation and decreased Cyclin A2 scripts would be those required for the exit from quies- and Cyclin E. Fmrp may directly target cell cycle tran- cence. A detailed understanding of the direct and scripts, blocking their translation in quiescence, but sta- indirect targets of Fmrp and Dcp1a in different cellular bilizing them for mobilization during reactivation. In states is needed to resolve this issue. absence of Fmrp during cell cycle entry, these transcripts may instead be targeted for translation and turnover, by the increase in Dcp1a. The phenotype of Dcp1a knock- Reciprocal effects of Fmrp and Dcp1a on the cell down during cell cycle re-entry from G0 was similar to cycle may reflect the balance between mRNA that of cycling myoblasts: increased EdU incorporation turnover and translation in control of cell state accompanied by increased Cyclin A2 and Cyclin E ex- Perturbing Fmrp and Dcp1a expression in proliferating pression. As mRNP puncta are assembled in cell cycle cells had contrasting impacts on the cell cycle. In cycling activated myoblasts within 3 h, the Dcp1a knockdown cells, Fmrp knockdown led to an increase in Dcp1a and may accelerate proliferation via increased accumulation a simultaneous reduction in EdU incorporation, suggest- of pro-cell cycle transcripts and increased translation, ra- ing that increased nonsense-mediated decay may lead to ther than sequestration. degradation of target mRNAs, and compromise S phase All cellular states express some transcripts with ex- entry. Support for this hypothesis comes from the obser- tremely short half-lives, and likely targets of Dcp1a. The vation that expression of Cyclin E, a key positive regula- work of the Coller group [52] has shown that genome- tor of the G1/S transition is suppressed in the Fmrp wide half-lives of transcripts are increased during quies- knockdown. Given the enrichment of Fmrp stalling cence in human fibroblasts, but regulators of this change complexes and the severe translational block in G0, a re- were not identified. The targets of Dcp1a in different quirement for Fmrp in sustaining expression of Cyclin E cellular states are likely to be numerous and context- may appear paradoxical. However, it is also possible that specific, as suggested by studies in oocytes and embryos diminished expression of Cyclin E reflects increased (Ma et al (2013); Eulalio et al, (2007) [53, 54]. In our Roy et al. Skeletal Muscle (2021) 11:18 Page 25 of 28 study, we show that transcripts of cyclins A2, B1, D1, E observations may be explained by a model wherein the and Ki67 are increased when Dcp1a expression is com- translation repression normally effected by Fmrp on tar- promised in G0, providing possible direct targets. The get mRNAs in proliferating myoblasts would be lifted in model in Fig. 10 outlines the possible targets in other the Fmrp knockdown, with consequent increase in states, suggesting reduced turnover and translation of Dcp1a-associated NMD complex resulting in possible pro-proliferative transcripts during quiescence and pro- degradation of transcripts including cyclins. The knock- myogenic differentiation respectively in Dcp1a-depleted down of Dcp1a could also lead to a decrease in the cells. By contrast, non-translating mRNAs held in re- ARE-mediated decay pathway [54, 55] leading to in- pressive mRNP granule complexes containing Fmrp creased half-life of cyclin and cytokine transcripts, po- would be expected to become targets for increased turn- tentiating cell cycle progression by preventing entry into over under conditions where Dcp1a is induced by G0 [26, 28]. Conceivably, altering the flux of different knockdown of Fmrp. Considering the compromised pro- transcripts through distinct puncta could alter the pro- liferation of Fmrp knockdown cells, we hypothesize that file of proteins synthesized, impacting proliferation. pro-proliferative transcripts are likely targets for rapid The molecular mechanism for reciprocal regulation turnover, as depicted in the model in Fig. 10. Indeed, that we observe in Fmrp and Dcp1a knockdown muscle our finding that mRNAs for Cyclins D, E A, and B all cells may involve direct mRNA binding by each protein, show decreased levels in the Fmrp knockdown and in- or may be mediated by indirect regulation of upstream creased levels in the Dcp1a knockdown (Fig. S7) is con- regulators. However, as knockdown of Fmrp and Dcp1a sistent with this scenario. each have broad phenotypic consequences for the cell cycle and differentiation, it is also possible that the al- tered levels of each protein (in the context of knock- Both Fmrp and Dcp1a are necessary for normal down of the other) are associated with the altered differentiation cellular state. At present, we cannot distinguish whether Whereas Fmrp and Dcp1a have opposing effects on pro- the mechanisms involving mRNP granules we describe liferation consistent with their opposing functions in are directly responsible for regulating MuSC fate, or a mRNA turnover vs. translation, differentiation is sup- consequence of signaling that affects global translation pressed when either Fmrp or Dcp1a are perturbed. Al- and that consequently impacts mRNP granules. though the direct targets are currently unknown, the Overall, these observations indicate that both Dcp1a mechanisms by which these two regulators affect myo- and Fmrp may play a role in the assembly of mRNP genesis are likely to differ. As Fmrp knockdown leads to complexes, and that individually their knockdown affects reduced proliferative capacity, reduced differentiation the expression of transcripts encoding other mRNP pro- may reflect the reduced number of cells available for teins. Dcp1a knockdown had more pronounced effects myogenic commitment. It has been reported [23] that on transcript abundance than Fmrp knockdown, consist- -/- Fmr1 MuSCs show lower accumulation of MyoD and ent with the expected differential effects of mRNA deg- Myf5 proteins through translational silencing, delaying radation versus translational stalling. Thus, Dcp1a entry into the differentiation program. Our study sug- knockdown, by altering transcript levels of Fmr1 a key gests that this effect could be at multiple regulatory mRNP player in G0-inducing conditions, may alter the nodes where Fmrp either directly or indirectly partici- equilibrium between mRNA decay and sequestration re- pates in decisions regarding cell fate. The effect of Dcp1a quired for achieving and maintaining the quiescent state. knockdown on differentiation is consistent with the ob- Our studies point to integrative mechanisms regulating served increase in proliferation, the antagonistic nature a critical balance between the mRNA decay and transla- of these programs being well reported. At a mechanistic tional repression, which enables expression of cell cycle level, the loss of differentiation potential may reflect the (and other) regulators that control proliferation, strong reduction of Myogenin protein. quiescence or differentiation. In summary, our results support a model where distinct mRNP constellations Cross regulation of mRNP granule components revealed characterize different cellular states and suggest that re- by knockdown analysis modeling these complexes may contribute to the transi- Knockdown of Fmrp resulted in an increase in Dcp1a tions between states. puncta, and knockdown of Dcp1a led to an increase of Abbreviations Fmrp in puncta (Fig. 7b) suggesting a reciprocal balance Ago2: Argonaute 2; CD45: Cluster of Differentiation 45; Dcp1a: Decapping between mRNA decay and translational repression. Spe- Protein 1a; Edc3: Enhancer of Decapping 3; Edc4: Enhancer of Decapping 4; EdU: 5-Ethynyl-2´-deoxyuridine; eIF-4E: Eukaryotic initiation factor 4E; eIF- cifically, our results point to a regulatory loop where 4F: Eukaryotic initiation factor 4F; Fmrp: Fragile X Mental Retardation Protein; Fmrp negatively regulates Dcp1a function and Dcp1a Ki67: Marker of Proliferation; LSm4: U6 snRNA-associated Sm-like protein negatively regulates Fmrp function (Fig. 10). Our LSm4; mRNP: Messenger Ribonucleoprotein; MuSC: Muscle Stem Cells; Roy et al. Skeletal Muscle (2021) 11:18 Page 26 of 28 Myf5: Myogenic factor 5; MyoD: Myogenic Determination Protein 1; OPP: O- Myogenin is exclusively expressed by MT, Cyclin D1 is enriched in MB, Propargyl Puromycin; P0: 60S acidic ribosomal protein P0; Pat1: DNA and G0 cells lack both proteins, confirming their quiescence by the ab- topoisomerase 2-associated protein PAT1; Pax7 : Paired box 7; PFA: Para- sence of both proliferation and differentiation programs. (B) Global Tran- formaldehyde; PB: Processing bodies; Scr: Scrambled; SG: Stress granules; script status: The global transcript levels of Cyclin D1, p27, MyoD1, Myf5, TA: Tibialis anterior; TIA1: Tia1 cytotoxic granule-associated RNA binding pro- Gapdh in MB, G0, MT were analysed using data sets from Venugopal tein; VCAM-1: Vascular cell adhesion molecule 1 et.al., [49] 2020. GEO database (Series GSE110742). *p < 0.05. **p < 0.01, ***p <0.001. Figure S6. Dcp1a and Fmrp Knockdowns have opposing ef- fects on cell proliferation but compromise differentiation. This data repre- sents quantification of western blots shown in Figure 8. (A to D): Supplementary Information Densitometry of western blots of Dcp1a, Fmrp , Cyclin A2 , Cyclin E pro- The online version contains supplementary material available at https://doi. teins normalized with Gapdh as internal control in MB , MT, G0 and R3 org/10.1186/s13395-021-00270-9. (E): Western blots of Myogenin and Myosin Heavy chain proteins in G0 and R3 with Gapdh as internal control. All bar graphs represent mean ± Additional file 1: Table 1. Bio-informatic analysis of transcripts encod- sd from n ≥ 3 ,Two tailed paired Student’s t-test is indicated is indicated ing mRNP components. To assess whether changes in expression of as *p < 0.05. *p < 0.01, ***p <0.001. Figure S7. Altered expression of cell mRNP proteins resulted from changes in expression of their mRNAs, we cycle and myogenic transcripts in knockdown cells held in G0-inducing con- used the recent RNA seq analysis derived from muscle satellite cells fixed ditions. Cells were transfected with siFmr1 and siDcp1a pools for 18 by perfusion of adult mice (to prevent cell activation that results from dis- hours, then placed in suspension culture and 48 hours later RNA was iso- ruption of the niche during isolation [42]. These fixed satellite cells are lated for qRT-PCR of Dcp1a, Fmr1, Ki67, Cyclin A2, Cyclin D1, Cyclin E1, thought to more accurately represent the quiescent (G0 state) and have Cyclin B, p27, p21, MyoD1, and Myf 5 Loss of Dcp1a leads to increased a transcriptome profile distinct from MuSC isolated without fixation, abundance of transcripts encoding positive regulators of the cell cycle which are now understood to represent cells in an early activation state. (Ki67, Cyclins), along with suppression of Cdk inhibitor p21 mRNA levels, Activated satellite cells (ASC) represent proliferating primary myoblasts 2.5 consistent with increased EdU incorporation. Transcripts encoding Pax7 days post isolation from the animal. Transcripts encoding P body genes and MyoD were suppressed. Knockdown of Fmr1 leads to reduced abun- were selected from the RNAseq data and grouped according to their dance of Cyclin A2, B and D1 mRNAs, consistent with decreased EdU in- function as outlined [43, 44]. We calculated fold changes from FPKM corporation. Gapdh was used as internal control and normalized to Scr values (Fragments Per Kilobase of transcript per Million mapped reads) G0 condition. All bar graphs represent mean ± sd from n =3. Two tailed RNA seq data comparing fixed (quiescent) satellite cells and activated sat- paired Student’s t-test is indicated is indicated as *p < 0.05. **p < 0.01, ellite cells [44] and used a cut-off of 1.5 +/- (for up regulation and down ***p <0.001. regulation). False Discovery Rate approach: Two stage step-up method of Benjamini, Krieger and Yekutieli was used and 10% FDR was set up for generating p values for the analysis. Figure S1. Differential association of decay complex proteins in different cellular states. Immuno-staining of Acknowledgements Dcp1a/Edc4/Pat1 (left) and Dcp1a/Ago2 (right) in muscle cells in culture: We thank Debarya Saha (CCMB) for help with global transcript analysis and quiescent (G0), 3 hr reactivated (R3), proliferative (MB), and differentiated Louise Moyle (KCL) for help with single myofibers. We gratefully (MT). Blue arrows indicate co-localization of Dcp1a/Edc4/Pat1 in puncta. acknowledge Colin Crist (McGill Univ.) for detailed discussions and sharing Red arrows indicate co-localization of Dcp1a/Ago2 in puncta. Note the data before publication, U Varshney (IISc), R Muddashetty and Vishal Tiwari absence of Dcp1a or Pat1 puncta in G0, and the rapid reassembly in R3. (InStem) for help with polysome profiling and S Chattarji (NCBS) for access to Also note prominent nuclear staining for Edc4 in G0. Figure S2. (A) Sup- Fmr1 knockout mouse tissue used in this study. We thank J Joseph (NCCS) plementary to Figure 4A Additional representative immunofluorescence for Dcp1a antibody. We are grateful for access to flow cytometry and images showing Fmrp (green) and Dcp1a (red) puncta in G0, MB and MT, imaging facilities (CIFF at NCBS-InStem and AIF at CCMB), as well as use of as well as cells reactivated for 3 hr from G0 (R3). Arrows indicate promin- the National Mouse Resource animal facility at NCBS-InStem and the Labora- ent puncta. (B). Corrected Mean Fluorescence intensities (CMI) of Fmrp tory Animal Facility at CCMB. and Dcp1a respectively in MB, G0, R3 and MT. For quantification, more than 3 cells per group was used and CMI intensities from more than 12 puncta were analysed. Corrected mean intensity was calculated using Authors’ contributions CMI= Total intensity of signal – (Area of signal x Mean background sig- JD, NR, FPS, MP, and SS conceived and designed the study. NR, FPS, MP, SG, nal). For quantification, more than 3 cells per group was used and MFI in- HG, AA, MR, and SS performed experiments and collected data. SMH and PZ tensities from more than 10 puncta was analysed. Figure S3. Knockdown supported and supervised isolation and analysis of MuSC on single mouse of Fmrp leads to reduced cell renewability and is not accompanied by apop- myofibers by FPS. JD, NR, MP, FPS, SMH, and SS wrote the paper. The tosis. (A) Colony formation assay shows that reduced EdU incorporation authors read and approved the final manuscript. in Fmrp knockdown cells correlates with compromised self-renewal. Bar graph represents mean ± sd from n=3 biological replicates. Two tailed paired Student’s t-test is indicated as ***p<0.001. (B) Proliferating myo- Funding blasts (MB) were treated with siRNAs (Scr, siFmr1) for 18 hr and harvested This work was supported by postdoctoral fellowships from the Govt. of India at 24 hrs for FACS analysis of 10,000 cells performed after staining for Department of Biotechnology (DBT) to NR, FPS and MP, graduate fellowships apoptosis markers. X-axis represents Annexin V and Y-axis represents pro- from the Council of Scientific and Industrial Research to MR, HG, and SS, and pidium iodide. Upper Panel: Flow cytometric profile for MB. Lower Panel: from the Tata Institute of Fundamental Research to AA, and grants from the Quantification of FACS plots shows that Fmrp knockdown cells do not Indo-Australia Biotechnology Fund (DBT) and the Indo-Danish Strategic Re- undergo apoptosis. n=3, mean ± sd. Figure S4. Knockdown of Fmrp is search Fund (DBT) to JD. Animal experiments were carried out at the Na- not accompanied by senescence. (A). SA β-galactosidase assay performed tional Mouse Resource in NCBS-InStem (partially funded by DBT grant BT/ in MB cells treated with siRNAs (Scr, siFmr1) for 24 h or reactivated from PR5981/MED/31/181/2012). SMH is a Medical Research Council Scientist with quiescence for proliferation for 24 hrs (R24) does not show any significant Programme Grant (G1001029 and MR/N021231/1) support. The lab of P.S.Z. is difference in X-gal staining between control and Fmr1 Knock down cells. supported in this work by grants from the Medical Research Council (MR/ (B). Analysis of DNA damage-induced foci of γH2AX in cells reactivated P023215/1 and MR/S002472/1). The collaborative work was supported by a from quiescence for 24h (R24) does not reveal any increase in Fmr1 BBSRC Partnership Award to King’s College London and InStem. knock down cells. (C). qRT-PCR analysis for p21 did not reveal any signifi- cant change in Fmr1 knocked down condition in MB, G0 or R24. All bar graphs represent mean ± sd from n =2. Two tailed paired Student’s t-test Availability of data and materials is indicated as ***p <0.001. Figure S5. (A): Phenotyping of 3 cellular condi- All mouse strains, cell lines, antibodies, and siRNAs are available tions: A replicate blot is shown of the data depicted in Figure 4C. commercially. Roy et al. Skeletal Muscle (2021) 11:18 Page 27 of 28 Declarations 17. Crist CG, Montarras D, Buckingham M. Short article muscle satellite cells are primed for myogenesis but maintain quiescence with sequestration of Myf5 Ethics approval and consent to participate mRNA targeted by microRNA-31 in mRNP granules. STEM. 2012;11(1):118–26 All experiments were performed under appropriate institutional approvals. https://doi.org/10.1016/j.stem.2012.03.011. 18. Vogler TO, Wheeler JR, Nguyen D, Hughes MP, Britson KA, Lester E, et al. TDP-43 and RNA form amyloid-like myo-granules in regenerating muscle. Consent for publication Nature. 2018; https://doi.org/10.1038/s41586-018-0665-2. All authors have given consent for submission of this manuscript. 19. Ciaccio C, Fontana L, Milani D, Tabano S, Miozzo M, Esposito S. Fragile X syndrome: a review of clinical and molecular diagnoses. Ital J Pediatr. 2017; Competing interests 43(1):39 http://ijponline.biomedcentral.com/articles/10.1186/s13052-017-03 All authors declare no competing interests. 55-y. 20. Mazroui R. Trapping of messenger RNA by Fragile X mental retardation Author details protein into cytoplasmic granules induces translation repression. Hum Mol Institute for Stem Cell Science and Regenerative Medicine, Bangalore, India. Genet. 2002;11(24):3007–17. 2 3 Centre for Cellular and Molecular Biology, Hyderabad, India. National 21. Muddashetty RS, Kelić S, Gross C, Xu M, Bassell GJ. Dysregulated Center for Biological Sciences, Bangalore, India. Manipal Academy of Higher metabotropic glutamate receptor-dependent translation of AMPA receptor Education, Manipal, India. King’s College London, Randall Centre for Cell & and postsynaptic density-95 mRNAs at synapses in a mouse model of Molecular Biophysics, New Hunt’s House, Guy’s Campus, London, UK. fragile X syndrome. J Neurosci. 2007;27(20):5338–48. 22. Khandjian EW, Bardoni B, Corbin F, Sittler A, Giroux S, Heitz D, et al. Novel Received: 4 November 2020 Accepted: 28 April 2021 isoforms of the fragile X related protein FXR1P are expressed during myogenesis. 1998;7(13):2121–8. 23. Fujita R, Zismanov V, Jacob JM, Jamet S, Asiev K, Crist C. Fragile X mental retardation protein regulates skeletal muscle stem cell activity by regulating References the stability of Myf5 mRNA. Skelet Muscle. 2017;7(1):1–10. 1. Subramaniam S, Sreenivas P, Cheedipudi S, Reddy VR, Shashidhara LS, 24. Vasudevan S, Steitz JA. AU-rich-element-mediated upregulation of Chilukoti RK, et al. Distinct transcriptional networks in quiescent myoblasts: translation by FXR1 and Argonaute 2. Cell. 2007;128(6):1105–18. a role for Wnt signaling in reversible vs. irreversible arrest. PLoS One. 2013; 25. Brooks SA, Blackshear PJ. Tristetraprolin (TTP): interactions with mRNA and 8(6):e65097 https://doi.org/10.1371/journal.pone.0065097. proteins, and current thoughts on mechanisms of action. Biochim Biophys 2. Sebastian S, Sreenivas P, Sambasivan R, Cheedipudi S, Kandalla P, Pavlath Acta. 2013;1829(6–7):666–79. GK, et al. MLL5, a trithorax homolog, indirectly regulates H3K4 methylation, 26. Sachidanandan C, Sambasivan R, Dhawan J. Tristetraprolin and LPS- represses cyclin A2 expression, and promotes myogenic differentiation. Proc inducible CXC chemokine are rapidly induced in presumptive satellite cells Natl Acad Sci U S A. 2009;106(12):4719–24. in response to skeletal muscle injury. J Cell Sci. 2002;115(13):2701–12. 3. Cheedipudi S, Puri D, Saleh A, Gala HP, Rumman M, Pillai MS, et al. A fine 27. Hausburg MA, Doles JD, Clement SL, Cadwallader AB, Hall MN, Blackshear balance: epigenetic control of cellular quiescence by the tumor suppressor PJ, et al. Post-transcriptional regulation of satellite cell quiescence by TTP- PRDM2/RIZ at a bivalent domain in the cyclin a gene. Nucleic Acids Res. mediated mRNA decay. Elife. 2015;2015(4):1–18. 2015;43(13):6236–56. 28. Rodgers JT, King KY, Brett JO, Cromie MJ, Charville GW, Maguire KK. 4. Cheung TH, Rando TA. Molecular regulation of stem cell quiescence. Nat mTORC1 controls the adaptive transition of quiescent stem cells from G0 to Rev Mol Cell Biol. 2013;14(6):329–40. GAlert. Nature. 2014;510(7505):393–6 https://doi.org/10.1038/nature13255. 5. Dhawan J, Laxman S. Decoding the stem cell quiescence cycle - lessons 29. Sambasivan R, Gayraud-Morel B, Dumas G, Cimper C, Paisant S, Kelly RG, et from yeast for regenerative biology. J Cell Sci. 2015;128(24):4467–74. al. Distinct regulatory cascades govern extraocular and pharyngeal arch 6. Liu L, Cheung TH, Charville GW, Hurgo BMC, Leavitt T, Shih J, et al. muscle progenitor cell fates. Dev Cell. 2009;16(6):810–21. Chromatin modifications as determinants of muscle stem cell quiescence 30. Yan QJ, Asafo-Adjei PK, Arnold HM, Brown RE, Bauchwitz RP. A phenotypic and chronological aging. Cell Rep. 2013;4(1):189–204 https://doi.org/10.101 and molecular characterization of the fmr1-tm1Cgr fragile X mouse. Genes 6/j.celrep.2013.05.043. Brain Behav. 2004;3(6):337–59. 7. Dhawan J, Rando TA. Stem cells in postnatal myogenesis: molecular 31. Fukada SI, Higuchi S, Segawa M, Koda KI, Yamamoto Y, Tsujikawa K, et al. mechanisms of satellite cell quiescence, activation and replenishment. Purification and cell-surface marker characterization of quiescent satellite Trends Cell Biol. 2005;15(12):666–73 https://doi.org/10.1016/j.tcb.2005.10.007. cells from murine skeletal muscle by a novel monoclonal antibody. Exp Cell 8. Comai G, Tajbakhsh S. Molecular and cellular regulation of skeletal Res. 2004;296(2):245–55. myogenesis. Curr Top Dev Biol. 2014;110:1–73 https://doi.org/10.1016/B978- 32. Benecke BJ, Ben-Ze’ev A, Penman S. The regulation of RNA metabolism in 0-12-405943-6.00001-4 Elsevier Inc. suspended and reattached anchorage-dependent 3 T6 fibroblasts. J Cell 9. Zismanov V, Chichkov V, Colangelo V, Jamet S, Wang S, Syme A, et al. Physiol. 1980;103(2):247–54. Phosphorylation of eIF2?? Is a translational control mechanism regulating 33. Arora R, Rumman M, Venugopal N, Gala H, Dhawan J. Mimicking muscle muscle stem cell quiescence and self-renewal. Cell Stem Cell. 2016;18(1):79– stem cell quiescence in culture: methods for synchronization in reversible arrest. Methods Mol Biol. 2017;1556:283–302. 10. Buchan JR, Parker R. Eukaryotic stress granules : the ins and out of 34. Milasincic DJ, Dhawan J, Farmer SR. Anchorage-dependent control of translation. What are stress granules? Mol Cell. 2009;36(6):932. muscle-specific gene expression in C2C12 mouse myoblasts. In Vitro Cell 11. Buchan JR. mRNP granules. Assembly, function, and connections with Dev Biol Anim. 1996;32(2):90–9. disease. RNA Biol. 2014;11(8):1019–30. 35. Faye MD, Graber TE, Holcik M. Assessment of selective mRNA translation in 12. Ramaswami M, Taylor JP, Parker R. Altered ribostasis: RNA-protein granules mammalian cells by polysome profiling. J Vis Exp. 2014;92:1–8. in degenerative disorders. Cell. 2013;154:727–36 https://doi.org/10.1016/j. 36. Parker R, Sheth U. P bodies and the control of mRNA translation and cell.2013.07.038 Elsevier Inc. degradation. Mol Cell. 2007;25:635–46. 13. Brengues M. Movement of eukaryotic mRNAs between polysomes and 37. Vasudevan S, Tong Y, Steitz JA. Switching from repression to activation: cytoplasmic processing bodies. Science. 2005;310(5747):486–9 http://www. microRNAs can up-regulate translation. Science. 2007;318(5858):1931–4. sciencemag.org/cgi/doi/10.1126/science.1115791. 14. Beckham CJ, Parker R. P Bodies, stress granules, and viral life cycles. Cell 38. Darnell JC, Van Driesche SJ, Zhang C, Hung KYS, Mele A, Fraser CE, et al. Host Microbe. 2008;3(4):206–12. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function 15. Decker CJ, Parker R. P-Bodies and stress granules : possible roles in the and autism. Cell. 2011;146(2):247–61 https://doi.org/10.1016/j.cell.2011.06. control of translation and mRNA degradation. 2012;4(9):a012286 https://doi. 013. org/10.1101/cshperspect.a012286. 39. Luo Y, Shan G, Guo W, Smrt RD, Johnson EB, Li X, et al. Fragile x mental 16. Aziz A, Liu QC, Dilworth FJ. Regulating a master regulator: establishing retardation protein regulates proliferation and differentiation of adult neural tissue-specific gene expression in skeletal muscle. Epigenetics. 2010;5(8): stem/progenitor cells. PLoS Genet. 2010;6(4):e1000898 https://doi.org/10.13 691–5. 71/journal.pgen.1000898. Roy et al. Skeletal Muscle (2021) 11:18 Page 28 of 28 40. Laggerbauer B, Ostareck D, Keidel EM, Ostareck-Lederer A, Fischer U. Evidence that fragile X mental retardation protein is a negative regulator of translation. Hum Mol Genet. 2001;10(4):329–38. 41. Kedersha N, Stoecklin G, Ayodele M, Yacono P, Lykke-Andersen J, Fitzler MJ, et al. Stress granules and processing bodies are dynamically linked sites of mRNP remodeling. J Cell Biol. 2005;169(6):871–84. 42. Yue L, Wan R, Luan S, Zeng W, Cheung TH. Dek modulates global intron retention during muscle stem cell quiescence exit. Dev Cell. 2020;53(6):661– 676.e6 https://doi.org/10.1016/j.devcel.2020.05.006. 43. Zheng D, Chen CYA, Shyu A. Bin. Unraveling regulation and new components of human P-bodies through a protein interaction framework and experimental validation. Rna. 2011;17(9):1619–1634. 44. Vanderweyde T, Apicco DJ, Youmans-Kidder K, Ash PEA, Cook C, Lummertz da Rocha E, et al. Interaction of tau with the RNA-binding protein TIA1 regulates tau pathophysiology and Toxicity. Cell Rep. 2016;15(7):1455–66. 45. Moon SL, Morisaki T, Khong A, Lyon K, Parker R, Stasevich TJ. Multicolour single-molecule tracking of mRNA interactions with RNP granules. Nat Cell Biol. 2019;21(2):162–8. 46. Ben-Ze’ev A, Farmer SR, Penman S. Mechanisms of regulating tubulin synthesis in cultured mammalian cells. Cell. 1979;17(2):319–25. 47. Fukada S, Uezumi A, Ikemoto M, Masuda S, Segawa M, Tanimura N, et al. Molecular signature of quiescent satellite cells in adult skeletal muscle. Stem Cells. 2007;25(10):2448–59. 48. Seale P, Sabourin LA, Girgis-Gabardo A, Mansouri A, Gruss P, Rudnicki MA. Pax7 is required for the specification of myogenic satellite cells. Cell. 2000; 102(6):777–86. 49. Venugopal N, Ghosh A, Gala H, Aloysius A, Vyas N, Dhawan J. The primary cilium dampens proliferative signaling and represses a G2/M transcriptional network in quiescent myoblasts. BMC Mol Cell Biol. 2020;21(1):25. 50. Patel PH, Barbee SA, Blankenship JT. GW-bodies and P-bodies constitute two separate pools of sequestered non-translating RNAs. PLoS One. 2016; 11(3):1–23. 51. Eulalio A, Behm-Ansmant I, Izaurralde E. P bodies: at the crossroads of post- transcriptional pathways. Nat Rev Mol Cell Biol. 2007;8(1):9–22. 52. Johnson EL, Robinson DG, Coller HA. Widespread changes in mRNA stability contribute to quiescence-specific gene expression patterns in a fibroblast model of quiescence. BMC Genomics. 2017;18(1):123 http://bmcgenomics. biomedcentral.com/articles/10.1186/s12864-017-3521-0. 53. Ma J, Flemr M, Strnad H, Svoboda P, Schultz RM. Maternally recruited DCP1A and DCP2 contribute to messenger RNA degradation during oocyte maturation and genome activation in Mouse1. Biol Reprod. 2013;88(1):1–12 https://academic.oup.com/biolreprod/article-lookup/doi/10.1095/ biolreprod.112.105312. 54. Eulalio A, Rehwinkel J, Stricker M, Huntzinger E, Yang SF, Doerks T, et al. Target-specific requirements for enhancers of decapping in miRNA- mediated gene silencing. Genes Dev. 2007;21(20):2558–70. 55. Stoecklin G, Mayo T, Anderson P. ARE-mRNA degradation requires the 5′ -3′ decay pathway. EMBO Rep. 2006;7(1):72–7. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Journal

Skeletal MuscleSpringer Journals

Published: Jul 8, 2021

Keywords: Quiescence; mRNP granule; Translational control; mRNA decay; Skeletal muscle; Myoblast; G0; Fmrp; Dcp1a; Fmr1 knockout; Muscle stem cell

References