LncRNA OIP5-AS1-directed miR-7 degradation promotes MYMX production during human myogenesis

Abstract

Long noncoding RNAs (lncRNAs) and microRNAs (miRNAs) modulate gene expression programs in physiology and disease. Here, we report a noncoding RNA regulatory network that modulates myoblast fusion into multinucleated myotubes, a process that occurs during muscle development and muscle regeneration after injury. In early stages of human myogenesis, the levels of lncRNA OIP5-AS1 increased, while the levels of miR-7 decreased. Moreover, OIP5-AS1 bound and induced miR-7 decay via target RNA-directed miRNA decay; accordingly, loss of OIP5-AS1 attenuated, while antagonizing miR-7 accelerated, myotube formation. We found that the OIP5-AS1-mediated miR-7 degradation promoted myoblast fusion, as it derepressed the miR-7 target MYMX mRNA, which encodes the fusogenic protein myomixer (MYMX). Remarkably, an oligonucleotide site blocker interfered with the OIP5-AS1-directed miR-7 degradation, allowing miR-7 to accumulate, lowering MYMX production and suppressing myotube formation. These results highlight a mechanism whereby lncRNA OIP5-AS1-mediated miR-7 decay promotes myotube formation by stimulating a myogenic fusion program.

Figure 1.

OIP5-AS1 binds and directs miR-7 degradation during myogenesis. (A) Fluorescent micrographs of MYH signals (red) to monitor the progression of human AB678 myoblasts from an undifferentiated state (day 0), when MYH is undetectable, to a state of full differentiation into myotubes (day 4), when MYH fluorescence is robust. Staining with DAPI (blue) was used to identify nuclei. (B) At the times indicated in differentiating AB678 cultures, the levels of myogenic proteins were assessed by western blot analysis. Bands were quantified by densitometry and the relative intensities are shown. (C) Complementarity between OIP5-AS1 and miR-7. At the times indicated in differentiating AB678 cultures, the relative levels of OIP5-AS1 (D), miR-7 (miR-7-5p) (E), miR-7-3p (F) and precursor (pre)-miR-7 (G) were quantified by RT-qPCR analysis. (H) Biotinylated Ctrl and miR-7 were transfected into AB678 cells; 24 h after adding differentiation medium, RNA complexes were pulled down using streptavidin beads (see the ‘Materials and Methods’ section) and the presence of OIP5-AS1 in the pulldown material was assessed by RT-qPCR analysis and normalized to the levels of GAPDH mRNA. (I) AB678 myoblasts were transfected with either Ctrl miR or miR-7 mimic; 24 h later, they were placed in differentiation medium for an additional 24 h, whereupon the levels of OIP5-AS1 were measured by RT-qPCR analysis. (J) AB678 myoblasts were transfected with Ctrl siRNA or OIP5-AS1-directed siRNA; 24 h later, they were placed in differentiation medium and were collected at the times shown after the induction of differentiation. The levels of miR-7 (left), miR-7-3p (middle) and pre-miR-7 (right) were all measured by RT-qPCR analysis. (K) AB678 myoblasts were transfected with Ctrl siRNA or ZSWIM8-directed siRNA; 24 h later, they were placed in differentiation medium and were collected 24 h after inducing differentiation. The levels of miR-7 (left), miR-7-3p (middle) and pre-miR-7 (right) were measured by RT-qPCR analysis. Data in panels (B) and (D)–(K) display the means ± SEM from three or more biological replicates. Significance was established using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. Other data are representative of three or more biological replicates.

Figure 2.

Silencing OIP5-AS1 attenuates myogenesis and inhibiting miR-7 promotes myogenesis. (A) AB678 myoblasts were transfected with Ctrl siRNA or OIP5-AS1-directed siRNA; 24 h later, they were placed in differentiation medium for 72 h, whereupon differentiation was monitored by assessing MYH levels (green) by immunofluorescence (left), and the fusion index and number of nuclei per myotube were quantified (right) after assessing five separate fields per experiment. (B) AB678 myoblasts were transfected with Ctrl siRNA or OIP5-AS1-directed siRNA; they were then placed in differentiation medium, and collected at the times shown after the induction of differentiation. The levels of MYOG, MYH and loading control HSP90 were assessed by western blot analysis. (C) AB678 myoblasts were transfected with Ctrl miR, miR-7 mimic or miR-7 antagomiR; they were then placed in differentiation medium for 72 h, and differentiation was monitored by assessing MYH levels (red) by immunofluorescence (left), and the fusion index and number of nuclei per myotube were quantified (right) after assessing five separate fields per experiment. (D) AB678 myoblasts were transfected with Ctrl miR, miR-7 mimic or miR-7 antagomiR; as described in panel (B), the levels of MYOG, MYH, MEF2C and loading control HSP90 were assessed by western blot analysis. (E) Cells were processed as described in panel (C), and the levels of CK activity were measured enzymatically (see the ‘Materials and Methods’ section). Data in all panels represent the means ± SEM from three or more biological replicates. Significance was established using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. In panels (B) and (D), bands were quantified by densitometry and the relative intensities are indicated.

Figure 3.

Antagonization of miR-7 rescues the anti-myogenic effect of silencing OIP5-AS1. (A) AB678 myoblasts were transfected with Ctrl siRNA or OIP5-AS1-directed siRNA together with Ctrl antagomiR or miR-7 antagomiR; they were then placed in differentiation medium, and collected at the times shown after the induction of differentiation. Differentiation was monitored by assessing MYH levels and loading control HSP90 by using western blot analysis. Bands were quantified by densitometry and the relative intensities are indicated. As described in panel (A), after differentiating for 72 h, myogenic progression was monitored by immunofluorescence assessment of MYH levels (B), by quantifying numbers of nuclei per myotube and fusion indices (five separate fields per experiment) (C), and by measuring the levels of CK activity (D). Data in panels (A), (C) and (D) are the means ± SEM from three or more biological replicates. Significance was established using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. In panel (A), bands were quantified by densitometry and the relative intensities are indicated.

Figure 4.

MYMX is a novel target of miR-7. (A) Venn diagram of RBPs from the top 100 upregulated RNAs in myogenesis by RNA-seq analysis (see the ‘Materials and Methods’ section) (gray), and the 875 predicted mRNA targets of miR-7 (orange); one mRNA was found in the intersection, MYMX mRNA. (B) AB678 myoblasts were transfected with biotinylated Ctrl miR or biotinylated miR-7; 6 h later, they were placed in differentiation medium for 12 h, harvested and RNA complexes were pulled down using streptavidin beads (see the ‘Materials and Methods’ section). The presence of potential miR-7 target mRNAs in the pulldown material was assessed by RT-qPCR analysis. The levels of RNAs tested in the pulldown were normalized to the levels of GAPDH mRNA. At the times indicated in differentiating AB678 cultures, the relative levels of MYMX mRNA and myogenic mRNAs were detected by RT-qPCR analysis (C), and the levels of MYMX and myogenic proteins were assessed by western blot analysis (D). (E) AB678 myoblasts were transfected with Ctrl siRNA or MYMX-directed siRNA; 24 h later, they were placed in differentiation medium for 72 h, and differentiation was monitored by assessing MYH levels by immunofluorescence (top), and the numbers of nuclei per myotube and fusion indices were quantified after assessing five separate fields per experiment (bottom). (F) As described in panel (E), the levels of MYMX and myogenic proteins were assessed by western blot analysis. Data in panels (B) and (D)–(F) represent the means ± SEM from three or more independent experiments. Significance was established using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. Other data are representative of three or more biological replicates. In panels (D) and (F), bands were quantified by densitometry and the relative intensities are indicated.

Figure 5.

miR-7 post-transcriptionally suppresses the stability and translation of MYMX mRNA. (A) AB678 myoblasts were transfected with Ctrl miR or miR-7 mimic; 24 h later, they were placed in differentiation medium, and collected at the times shown after the induction of differentiation. The levels of MYMX mRNA were measured by RT-qPCR analysis. (B) AB678 myoblasts were transfected with Ctrl miR, miR-7 mimic as described in panel (A) or miR-7 antagomiR, and MYMX protein levels were assessed by western blot analysis. (C) Schematic of the psiCHECK2 dual-luciferase system, which contains RL coding region fused to the WT MYMX 3′UTR [psiCHECK2-MYMX(3′WT)] or the MYMX 3′UTR bearing a mutation in the miR-7 binding site [psiCHECK2-MYMX(3′mut)]. FL activity, expressed from the same vector, was measured as an internal control. (D, E) Twelve hours after co-transfecting myoblasts with the parent reporter (psiCHECK2) or test reporter psiCHECK2-MYMX(3′WT) with either Ctrl miR or miR-7 into AB678 myoblasts and inducing differentiation, RL/FL activity was calculated. Twenty-four hours after co-transfecting the plasmids psiCHECK2-MYMX(3′WT) and psiCHECK2-MYMX(3′mut), AB678 myoblasts were induced to differentiate for 24 h, and the relative RL/FL ratios were determined. (F) Twenty-four hours after co-transfecting psiCHECK2 or test reporter psiCHECK2-MYMX(3′WT) and control or OIP5-AS1-directed siRNA, AB678 myoblasts were induced to differentiate for 24 h, and the relative RL/FL ratios were determined. (G) Twenty-four hours after transfecting Ctrl miR-7 or miR-7 mimic, AB678 myoblasts were placed in differentiation medium for an additional 24 h; cells were then treated with actinomycin D, and the relative levels of MYMX mRNA (left) and a control stable transcript, GAPDH mRNA (right), were assessed by RT-qPCR analysis and normalized to 18S rRNA levels, also quantified by RT-qPCR analysis. mRNA half-lives (t_1/2) were calculated as the times required to reach 50% of the initial abundance of the mRNA at time 0 before adding actinomycin D. (H) Twenty-four hours after transfecting Ctrl miR or miR-7 mimic, AB678 myoblasts were placed in differentiation medium for an additional 24 h, and then harvested. Cytoplasmic extracts were fractionated by centrifugation through sucrose density gradients, RNA was isolated from equal volumes of each fraction and the abundance of MYMX and ACTB mRNAs was quantified by RT-qPCR analysis and represented as a percentage of the total mRNA on the gradient. AB678 myoblasts were co-transfected with Ctrl miR or miR-7 mimic and with either a control empty vector (EV) plasmid or a plasmid expressing myc-tagged MYMX; 24 h later, AB678 myoblasts were induced to differentiate and 48 h after that the levels of MYMX, MYMX-myc, and HSP90 were determined by western blot analysis (I), differentiation was monitored by assessing MYH immunofluorescence (J) and the fusion index and the number of nuclei per myotube were quantified; five fields were assessed per experiment (K). Data in panels (A), (B), (D)–(G), (I) and (K) represent the means ± SEM from at least three independent experiments. Significance was established using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. Other data are representative of three or more biological replicates.

Figure 6.

TDMD of miR-7 by OIP5-AS1 affects myoblast fusion. (A) Schematic of fusion assays. AB678 cells were labeled by expression of enhanced green fluorescent protein (EGFP) or red fluorescent protein mCherry following lentivirus infection. Further silencing of OIP5-AS1 or overexpression of miR-7 or corresponding controls was carried out on EGFP-labeled AB678 cells. Equal numbers of EGFP-labeled or mCherry-labeled AB678 cells were then mixed and placed in differentiation medium for 72 h. (B) As described in panel (A), EGFP-labeled AB678 cells were transfected with Ctrl miR (left) or miR-7 mimic (right) and further mixed with mCherry-labeled AB678. The fusion ability was monitored by representative confocal images, homologous fusion (EGFP+ only or mCherry+ only) and heterologous syncytia (both EGFP+ and mCherry+). (C) As described in panel (A), EGFP-labeled AB678 cells were transfected with Ctrl siRNA (left) or OIP5-AS1-directed siRNA (right) and further mixed with mCherry-labeled AB678. The fusion ability was monitored by representative confocal images of homologous fusion (EGFP+ only or mCherry+ only) and heterologous syncytia (both EGFP+ and mCherry+). (D, E) Heterologous fusion indices were quantified for panels (B) and (C) after assessing five separate fields per experiment. Scale bar: 100 μm. Data in panels (D) and (E) represent the means ± SEM from at least three independent experiments. Significance was established using Student’s t-test. ***P < 0.001.

Figure 7.

TSB of OIP5-AS1:miR-7 attenuates myotube formation by inhibiting OIP5-AS1-directed miR-7 degradation. (A) AB678 myoblasts were transfected with TSB (Ctrl TSB or OIP5-AS1:miR-7 TSB); 24 h after transfection, AB678 myoblasts were induced to differentiate, and were collected at the indicated times. The levels of miR-7 (left), miR-7-3p (middle) and pre-miR-7 (right) were measured by RT-qPCR analysis. (B) In cells prepared as described in panel (A), the levels of OIP5-AS1 were measured by RT-qPCR analysis. (C) AB678 myoblasts were transfected with Ctrl TSB or OIP5-AS1:miR-7 TSB; 24 h later, they were placed in differentiation medium for 72 h, and differentiation was monitored by assessing MYH levels by immunofluorescence, and the fusion indices and numbers of nuclei per myotube (D) were quantified after assessing five separate fields per experiment. (E) In cells prepared as described in panel (A), the levels of MYMX mRNA were measured by RT-qPCR analysis. (F) AB678 myoblasts were transfected with Ctrl TSB or OIP5-AS1:miR-7 TSB; 24 h later, they were placed in differentiation medium, and collected at the times shown after the induction of differentiation. The levels of MYMX, MYH and loading control HSP90 were assessed by western blot analysis. Bands were quantified by densitometry and the relative intensities are indicated. (G) Twenty-four hours after co-transfecting the plasmids bearing wild-type or mutant MYMX 3′UTR with Ctrl TSB or OIP5-AS1:miR-7 TSB, AB678 myoblasts were induced to differentiate for 24 h, whereupon RL/FL ratios were determined. As described in Figure 6, EGFP-labeled AB678 cells were transfected with Ctrl TSB (left) or OIP5-AS1:miR-7 TSB (right) and further mixed with mCherry-labeled AB678. The fusion ability was monitored by representative confocal images (H) showing homologous fusion (EGFP+ or mCherry+) and heterologous syncytia (both EGFP+ and mCherry+), and the heterologous fusion index (I) was quantified after assessing five separate fields per experiment. AB678 myoblasts were transfected with Ctrl TSB or OIP5-AS1:miR-7 TSB along with either Ctrl antagomiR or miR-7 antagomiR; they were then placed in differentiation medium for 72 h, and the progression of differentiation was monitored by assessing MYH levels by immunofluorescence (J), and by measuring fusion indices and numbers of nuclei per myotube (K) in five separate fields per experiment. (L) Expression levels of 37 myogenesis-associated mRNAs that are predicted targets of miR-7, are significantly downregulated after introducing OIP5-AS1:miR-7 TSB (by 0, 24 or 48 h of differentiation) and are upregulated during myogenesis for 24 h (see also Supplementary Figure S6). Significance was established using P_adj < 0.05, and log₂FC > 1 (Prolif: proliferating; Diff: differentiated). (M) GSEA of differentially expressed genes in OIP5-AS1:miR-7 TSB treated cells. GO annotations based on cellular components; top 10 gene sets were established using normalized enrichment score. (N) AB678 myoblasts were transfected with Ctrl TSB or OIP5-AS1:MEF2C TSB; 24 h later, they were placed in differentiation medium for 24 h, and the binding of HuR to MEF2C mRNA was assessed by cross-linking RIP analysis; data were normalized to the levels of GAPDH mRNA in each IP sample and represented as the enrichment of each mRNA in HuR IP samples relative to the levels of the mRNA in IgG IP samples. AB678 myoblasts were transfected with Ctrl TSB, OIP5-AS1:miR-7 TSB, OIP5-AS1:MEF2C TSB (50 nM each) or with a cocktail of OIP5-AS1:miR-7 TSB (25 nM) plus OIP5-AS1:MEF2C TSB (25 nM); 24 h later, they were placed in differentiation medium for 72 h, and differentiation was monitored by assessing MYH levels by immunofluorescence (O), and by measuring the fusion indices and the numbers of nuclei per myotube (P) in five separate fields per experiment. Data in panels (A), (B), (D)–(G), (I), (K), (N) and (P) represent the means ± SEM from at least three independent experiments. Significance was established using Student’s t-test. **P < 0.01; ***P < 0.001. Other data are representative of three or more biological replicates.

Figure 8.

Model. In proliferating (undifferentiated) myoblasts, lncRNA OIP5-AS1 levels are low, while miRNA miR-7 levels are high. As myogenic differentiation progresses, OIP5-AS1 levels rise, and through its extensively complementarity with miR-7, it induces miR-7 degradation via TDMD (right arm). The low levels of miR-7 in differentiated myoblasts then permit the rise in expression of the miR-7 target MYMX mRNA, which encodes the fusogenic protein MYMX and further promotes myotube fusion. Use of a TSB that masks the site of miR-7 binding on OIP5-AS1 (left arm) prevents miR-7 decay, in turn keeping MYMX mRNA unstable and repressing myogenesis. In sum, through a TDMD mechanism, OIP5-AS1 promotes fusogenic protein MYMX expression and enhances myotube formation.