pubmed.ncbi.nlm.nih.gov

The microprotein Minion controls cell fusion and muscle formation - PubMed

  • ️Sun Jan 01 2017

The microprotein Minion controls cell fusion and muscle formation

Qiao Zhang et al. Nat Commun. 2017.

Abstract

Although recent evidence has pointed to the existence of small open reading frame (smORF)-encoded microproteins in mammals, their function remains to be determined. Skeletal muscle development requires fusion of mononuclear progenitors to form multinucleated myotubes, a critical but poorly understood process. Here we report the identification of Minion (microprotein inducer of fusion), a smORF encoding an essential skeletal muscle specific microprotein. Myogenic progenitors lacking Minion differentiate normally but fail to form syncytial myotubes, and Minion-deficient mice die perinatally and demonstrate a marked reduction in fused muscle fibres. The fusogenic activity of Minion is conserved in the human orthologue, and co-expression of Minion and the transmembrane protein Myomaker is sufficient to induce cellular fusion accompanied by rapid cytoskeletal rearrangement, even in non-muscle cells. These findings establish Minion as a novel microprotein required for muscle development, and define a two-component programme for the induction of mammalian cell fusion. Moreover, these data also significantly expand the known functions of smORF-encoded microproteins.

PubMed Disclaimer

Conflict of interest statement

All listed authors are employees of the Novartis Institutes for BioMedical Research.

Figures

Figure 1
Figure 1. The microprotein Minion is specifically expressed during skeletal muscle development and regeneration.

(a) Left: Overlap of RNA-seq from regenerating adult mouse tibialis anterior (TA) muscle and differentiating C2C12 myoblasts (MB). CTX, cardiotoxin; FC, fold change compared to uninjured muscle (bottom left) or undifferentiated myoblasts (bottom right). Right: fold change of reads per kilobase per million mapped reads (RPKM) for selected genes upregulated after CTX injury. Values are normalized to uninjured muscle, representing mean±s.d. of fold change, three 8–10 week old mice per time point. (b) Western blot of control uninjured (Ctl) and CTX-injured regenerating adult TA muscle (n=2, two 8–10 week old mice per time point, two technical replicates each). (c) Western blot of embryonic muscle samples (n=3, two embryos each). Day 4 post-CTX TA or normal saline (NS) injection were positive and negative controls. E, embryonic day; P, post-natal. (d) Western blot of C2C12 myoblasts cultured under growth conditions (GM) or under differentiation conditions (DM) for the indicated number of days (n=4). MHC, myosin heavy chain. (b–d) Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) and Tubulin served as loading controls. (e) Protein sequence alignment of mouse Minion with putative orthologues from other mammalian species (GenBank accession numbers and UniProt IDs from top to bottom are: NP_001170939.1, XP_017452417.1, NP_001302423.1, EHH18375.1, M3X8W7_FELCA and F1RQU5_PIG.)

Figure 2
Figure 2. Minion is required for skeletal muscle development.

(a) Strategy for CRISPR/Cas9 mutagenesis of the gm7325/Minion locus using a dual sgRNA approach. Grey box, Minion ORF; white box, non-coding exons; sgRNA, single gRNA; Fwd and Rev, forward and reverse genotyping primers. (b) Left: representative genotyping PCR of Minion wild type (+/+) and heterozygous (Δ/+) mice carrying the 135-bp deletion depicted in a. n=20 (more than 300 total adult mice of both sex). Right: representative sequence traces. Black line indicates 5′ boundary of the deletion. WT, wild-type allele; KO, knockout allele (135-bp deletion). (c) Photographs of skinned Minion+/+ and MinionΔ/Δ P0 mice. Cyan arrows and Blue asterisks indicate forelimb and intercostal musculature, respectively. Three litters of embryos. (d) Histological and immunofluorescence (IF) analyses of embryonic tongue skeletal muscle from E18.5 embryos. Yellow arrowheads and yellow arrows indicate multinuclear myofibres and unfused differentiating elongating myoblasts, respectively. Top row: hematoxylin and eosin (H&E) staining of sagittal tongue sections. Inset demonstrates the originating region and orientation of the provided tongue sections. Bottom row: Immunofluorescence staining for the muscle marker MHC (red), with DNA counterstain DAPI (4′,6-diamidino-2-phenylindole; blue). Insets demonstrate magnification of the boxed areas. Three litters of embryos. (e) Histological images of H&E-stained E19.5 forelimb longitudinal sections of indicated genotypes. Two litters of embryos. (f) Immunofluorescence images of forelimb longitudinal sections for E19.5 embryos with indicated genotypes. MHC (red) and DAPI (blue) staining are shown. Two litters of embryos. (d–f) Paraffin-embedded embryos of different stages were examined. (c–f) Embryos from the same litter were compared with 1–2 embryos for each genotype in each experimental repeat. Scale bars, 1 mm (c), 100 μm (d–f).

Figure 3
Figure 3. Minion deficiency blocks non-somitic skeletal muscle formation.

(a) Immunofluorescence images of non-somitic facial musculature from sagittal sections of E18.5 embryos with the indicated genotypes. Desmin (red) staining is shown. Black box in histological image at left demonstrates the area shown at right in fluorescence images. (b) Immunofluorescence images of non-somitic jaw and facial musculature on transverse sections from E19.5 embryos with the indicated genotypes. MHC (red) and DAPI (blue) staining are shown. (a,b) Two litters of embryos were examined, and 1–2 embryos of each genotype and in the same litter were compared in each experimental repeat. Scale bars, 1 mm for H&E images and 500 μm for immunofluorescence images.

Figure 4
Figure 4. Minion loss severely impairs respiratory function.

(a) Immunofluorescence images of intercostal muscle sagittal sections for E19.5 embryos with indicated genotypes. Desmin (red) and DAPI (blue) staining are shown. n=2. (b) Immunofluorescence images of sagittal sections of diaphragm muscle from E19.5 embryos stained for the muscle marker Desmin (red) and DAPI (blue). n=2 (four different sections each). (a,b) Two litters of embryos were examined, and two embryos for each genotype in each experimental repeat. (c) Quantification (left) and representative image (right) of lung flotation assay using E18.5 mouse embryos (56 total) following 1 h exposure to room air after caesarean section. Scale bars, 200 μm (a, b).

Figure 5
Figure 5. Minion is specifically required for fusion of skeletal muscle progenitors.

(a) Immunofluorescence of primary embryonic myoblasts isolated from E18.5 Minion+/+, MinionΔ/+ and MinionΔ/Δ embryos, following 3 days in DM. Desmin (green) and DAPI (red). White arrowheads: myotubes; white arrows: elongating myoblasts. n=3 (five technical replicates each). (b) Fusion index of myoblasts in a, calculated as % nuclei in Desmin+ myotubes (≥3 nuclei) of total nuclei in Desmin+ cells. Asterisk: P<0.05, unpaired two-tailed Student's t-test. Each value represents mean±s.d. n=4 (two 0.7 mm × 0.7 mm fields each). (c) Western blots of C2C12 myoblasts cultured in GM or DM. Cells were lentivirally infected with either control luciferase targeting shRNA (Ctrl) or serially with two shRNA targeting the Minion 3′UTR (MinionKD) and cultured in GM or DM for the indicated number of days. n=3. (d) Immunofluorescence images of Ctrl and MinionKD myofibres following 5 days in DM. MHC (green) and DAPI (red). n=5 (eight technical replicates each). (e) Differentiation index for d, calculated as % nuclei in MHC+ cells of total nuclei. NS, not significant, unpaired two-tailed Student's t-test. (f) Fusion index for d, calculated as % nuclei in MHC+ myotubes (≥3 nuclei) of total nuclei. Double asterisks: P<0.001, unpaired two-tailed Student's t-test. (g) Quantification of myotubes by nuclei number for d. For e–g, each value represents mean±s.d. n=4 (two 0.7 mm × 0.7 mm fields each). (h) Immunofluorescence images of MinionKD cells expressing either control protein (NanoLuc) or human MINION orthologue, after 5 days in DM. MHC (green) and DAPI (red). n=3 (six technical replicates each). Scale bars, 100 μm.

Figure 6
Figure 6. Minion is required for Myomaker-mediated fusion.

(a) Immunofluorescence images of cell-mixing between 10T1/2 fibroblasts and wild-type C2C12 myoblasts (1:2 ratio) after 3 days in differentiation medium. Differentiating myoblasts and myotubes are marked by MHC (red). 10T1/2 fibroblasts were infected with retrovirus expressing GFP (green) and proteins of interest (left, NanoLuc control; middle, mouse Myomaker; right, mouse Minion). DAPI marks nuclei. Fibroblasts expressing Myomaker fused with wild-type differentiating myoblasts and myotubes to become large thick myotubes (white arrowheads), while fibroblasts expressing Minion failed to do so. White arrows indicate MHC-positive myotubes that are not fused to fibroblasts. n=2 (eight technical replicates each). 0.7 mm × 0.7 mm fields are shown. (b) Immunofluorescence images of cell mixing between 10T1/2 fibroblasts expressing Myomaker and C2C12 myoblasts after 3 days in DM. 0.7 mm × 0.7 mm fields at × 20 magnification are shown. Control and MinionKD myoblasts were used. Differentiating myoblasts and myotubes are marked by MHC (red). 10T1/2 fibroblasts expressing Myomaker were labelled with CellTrace Violet dye before mixing (pseudocoloured in green). Fibroblasts expressing Myomaker fused with differentiating control myoblasts (white arrowheads), but failed to fuse to MinionKD myoblasts. n=2 (eight technical replicates each). (c) Western blot of wild-type C2C12 myoblasts in GM, as well as Ctrl and MinionKD myoblasts in DM for 3 days. n=2. (d) Immunofluorescence images of MinionKD myoblasts expressing Luciferase (Ctrl), Minion, or Myomaker, after 5 days in DM. MHC (green) and DAPI (red). n=2 (seven technical replicates each). (e) Western blot of cell lines shown in d. n=2. Scale bars, 100 μm.

Figure 7
Figure 7. Minion and Myomaker are sufficient to induce fusion of non-muscle cells.

(a) Western blot of 10T1/2 fibroblasts expressing Luciferase (Ctrl), Myomaker or Minion. n=2. (b) Retroviral vectors encoding Luciferase, Myomaker or Minion were transduced as indicated into 10T1/2 fibroblasts. All vectors contain IRES-GFP downstream of the gene of interest, causing infected cells to uniformly express GFP. Split-channel grayscale images for GFP and DNA are included. n=3 (eight technical replicates each). (c) Quantification of GFP+ syncytia in fibroblasts expressing combinations of proteins as indicated. Fusion index was calculated as the percentage of nuclei found within GFP-positive syncytia containing ≥3 nuclei. Syncytia were scored 24 h after seeding. Each value represents mean±s.d. n=3 (two 1.4 mm × 1.4 mm fields each). Double asterisks: P<0.001, unpaired two-tailed Student's t-test. (d) Fluorescence images from cell-mixing experiments using differentially labelled 10T1/2 fibroblasts. Cells were serially infected with retroviruses encoding the indicated combinations of Minion, Myomaker, or controls (label omitted for simplicity). CellTrace Violet (cyan) and CellTracker DeepRed (red) dyes were used for labelling. Yellow arrows indicate syncytia containing both DeepRed+ cells and Violet+ nuclei. White arrows indicate syncytia derived from DeepRed+ cells only. n=5 (six technical replicates each). See Supplementary Fig. 19 for split-channel images. (e) Quantification of fusion in d (bottom panels), measured as percentage of DeepRed+ syncytia (≥3 nuclei) containing ≥1 Violet+ nucleus. Each value represents mean±s.d. n=4 (six 1.4 mm × 1.4 mm fields each). NS, not significant; single asterisk: P<0.05, unpaired two-tailed Student's t-test. Scale bars, 100 μm (b) and 50 μm (d).

Figure 8
Figure 8. Minion and Myomaker-induced fusion requires cytoskeleton reorganization.

(a) Fluorescence images of 10T1/2 fibroblasts co-overexpressing Minion and Myomaker. F-actin (Alexa546-Phalloidin, red) and DAPI (blue) staining are shown. White arrowheads point to the boundaries of multinuclear cells. n=2 (six technical replicates each, five fields each). (b) Fluorescence images of 10T1/2 fibroblasts co-overexpressing Minion and Myomaker and treated for 24 h with DMSO control or the actin polymerization inhibitors latrunculin B (0.1 μM) or cytochalasin D (0.3 μM). n=2 (six technical replicates each, five fields each). Scale bars, 100 μm (a,b). (c) A proposed model for Minion and Myomaker-induced cell–cell fusion. We suggest that Minion and Myomaker have separable roles in the fusion process; Myomaker mediates pre-fusion pore events such as cell-cell recognition and/or, membrane apposition, whereas Minion mediates later fusion pore formation, at least in part via induction of cytoskeletal rearrangements.

Similar articles

Cited by

References

    1. Ma J. et al.. Improved identification and analysis of small open reading frame encoded polypeptides. Anal. Chem. 88, 3967–3975 (2016). - PMC - PubMed
    1. Saghatelian A. & Couso J. P. Discovery and characterization of smORF-encoded bioactive polypeptides. Nat. Chem. Biol. 11, 909–916 (2015). - PMC - PubMed
    1. Anderson D. M. et al.. A micropeptide encoded by a putative long noncoding RNA regulates muscle performance. Cell 160, 595–606 (2015). - PMC - PubMed
    1. Anderson D. M. et al.. Widespread control of calcium signaling by a family of SERCA-inhibiting micropeptides. Sci. Signal. 9, ra119 (2016). - PMC - PubMed
    1. Magny E. G. et al.. Conserved regulation of cardiac calcium uptake by peptides encoded in small open reading frames. Science 341, 1116–1120 (2013). - PubMed

Publication types

MeSH terms

Substances