Human myofiber-enriched aging-induced lncRNA FRAIL1 promotes loss of skeletal muscle function

Abstract

The loss of skeletal muscle mass during aging is a significant health concern linked to adverse outcomes in older individuals. Understanding the molecular basis of age-related muscle loss is crucial for developing strategies to combat this debilitating condition. Long noncoding RNAs (lncRNAs) are a largely uncharacterized class of biomolecules that have been implicated in cellular homeostasis and dysfunction across a many tissues and cell types. To identify lncRNAs that might contribute to skeletal muscle aging, we screened for lncRNAs whose expression was altered in vastus lateralis muscle from older compared to young adults. We identified FRAIL1 as an aging-induced lncRNA with high abundance in human skeletal muscle. In healthy young and older adults, skeletal muscle FRAIL1 was increased with age in conjunction with lower muscle function. Forced expression of FRAIL1 in mouse tibialis anterior muscle elicits a dose-dependent reduction in skeletal muscle fiber size that is independent of changes in muscle fiber type. Furthermore, this reduction in muscle size is dependent on an intact region of FRAIL1 that is highly conserved across non-human primates. Unbiased transcriptional and proteomic profiling of the effects of FRAIL1 expression in mouse skeletal muscle revealed widespread changes in mRNA and protein abundance that recapitulate age-related changes in pathways and processes that are known to be altered in aging skeletal muscle. Taken together, these findings shed light on the intricate molecular mechanisms underlying skeletal muscle aging and implicate FRAIL1 in age-related skeletal muscle phenotypes.

Keywords: aging; lncRNA; skeletal muscle.

FIGURE 1

FRAIL1 is a highly abundant skeletal muscle lncRNA induced by aging. (a) Overview of the study design and strategy used for identification of differentially expressed human lncRNA transcripts during skeletal muscle aging. RNA‐sequencing data from a previous study were filtered with high stringency and analyzed for lncRNA expression in skeletal muscle (average normalized RNA‐Seq counts across all samples ≥10) and differential expression between aged and young samples. (b) MA plot (log‐intensity ratios (M values, y‐axis) versus log‐intensity averages (A values, x‐axis)) of differences in lncRNA abundance between aged and young samples and average lncRNA abundance across all samples. Gray symbols represent lncRNA transcripts that were expressed below the abundance cutoff, black symbols represent skeletal muscle‐expressed lncRNAs that did not change with age, red circles are increased transcripts, and blue circles are decreased transcripts (smaller circles p < 0.05; larger circles Benjamini‐Hochberg corrected FDR <0.1).

FIGURE 2

FRAIL1 is skeletal muscle‐enriched and associated with characteristics of aging muscle. (a) GTEx bulk tissue expression levels of FRAIL1 lncRNA expressed as the median TPM value for each human tissue type relative to the expression level in skeletal muscle. (b) Overview of human study participant characteristics and phenotypic measurements. (c) Relative skeletal muscle FRAIL1 lncRNA level across study participants. Data are organized by age group and sex and normalized to the average FRAIL1 lncRNA level in young males. p Values were determined by one‐way ANOVA. (d–f) Age‐ and sex‐specific muscle function outcomes and muscle mass estimates maximal knee extension muscle strength (d), peak knee extension power (e), leg lean mass (f), and skeletal muscle index (g) in young males (YM), young females (YF), older males (OM), and older females (OF). *p < 0.05, ****p < 0.0001.

FIGURE 3

The transcription factors RBPJ, TEAD3, MYOG, and FOXO3 potentially regulate FRAIL1 expression. (a) The JASPAR database was queried for transcription factor binding sites located within the promoter region 350 nucleotides upstream of the FRAIL1 transcription start site. (b) Human vastus lateralis abundance levels of mRNAs encoding the transcription regulators RBPJ, TEAD3, MYOG, and FOXO3. Data are organized by age group and sex. For a comprehensive list of identified transcription factor binding sites and expression levels, see Figure S3. Horizontal bars indicate mean values from each group, false discovery rate (FDR) was determined by Benjamini‐Hochberg correction, and p values were determined by one‐way ANOVA. *p < 0.05, **p < 0.001.

FIGURE 4

Forced expression of FRAIL1 in mouse skeletal muscle causes muscle atrophy. (a–e) One TA muscle per mouse was transfected with empty plasmid (pcDNA3), and the contralateral TA in each mouse was transfected with plasmid encoding FRAIL1 under the control of the cytomegalovirus (CMV) promoter, as indicated. Seven days posttransfection, bilateral TAs were harvested for RT‐qPCR analysis of FRAIL1 expression (a) histological analysis of skeletal muscle fiber size (b, c) and immunofluorescence microscopy using antibodies targeting MYH2, MYH1, MYH4, and laminin for the quantification of muscle fiber type distributions (d, e). (f) Schematic diagram of the FRAIL1 transcript and synthetic truncated isoforms. The region conserved between nonhuman primates is highlighted in green. (g–i) One TA muscle per mouse was transfected with 10 μg of empty plasmid, and the contralateral TA in each mouse was transfected with 10 μg of plasmid encoding one of the truncated forms of FRAIL1 under the control of the CMV promoter, as indicated. Seven days posttransfection, bilateral TAs were harvested for RT‐qPCR analysis of RNA expression (g) and histological analysis of skeletal muscle fiber size (h, i). Shades of green (c, i) represent differential muscle fiber plasmid uptake following the transfection procedure. Horizontal bars indicate mean values from each group ±SEM, and p values were determined by paired (b, e, h) and unpaired (g) two‐tailed t tests. *p < 0.05, **p < 0.01.

FIGURE 5

Forced FRAIL1 expression alters the global skeletal muscle proteome and transcriptome. (a, b) One TA muscle per mouse (n = 5) was transfected with 10 μg of empty plasmid (pcDNA3), and the contralateral TA in each mouse was transfected with 10 μg of full‐length FRAIL1 expressing plasmid. Seven days posttransfection, bilateral TAs were harvested and subjected to TMT‐mass spectrometry. In a separate cohort, one TA muscle per mouse (n = 6) was transfected with 10 μg of empty plasmid (pcDNA3), and the contralateral TA in each mouse was transfected with 10 μg of full‐length FRAIL1 expressing plasmid. Seven days posttransfection, bilateral TAs were harvested and subjected to RNA‐seq analysis. (a, b) Volcano plots of differences in global protein (a) and transcript (b) abundance following ectopic FRAIL1 expression. (c, d) Heatmaps of key skeletal muscle proteins (c) and mRNAs (d) differentially expressed in response to FRAIL1 expression. False discovery rate (FDR) was determined by Benjamini‐Hochberg correction. * FDR ≤ 0.05, ** FDR ≤ 0.01.

FIGURE 6

Forced FRAIL1 expression increases the abundance of proteins involved in RNA splicing and alters the splicing of skeletal muscle mRNAs. (a–c) Enrichment plots of overrepresented GO terms among proteins increased by ectopic FRAIL1 expression in mouse skeletal muscle for biological process (a), cellular component (b), and molecular function (c) categories. (d–f) Heatmaps of key upregulated proteins from the RNA splicing (d), spliceosomal complex (e), and mRNA‐binding (f) ontologies. (g) Diagram of the mouse Tnnt3 locus and relative expression of TNNT3 protein, Tnnt3 mRNA, and Tnnt3 exons 1, 2, 8, 9, and 19 between control and FRAIL1 expressing mouse skeletal muscles. (h) Diagram of the mouse Ndufv3 locus and relative expression of NDUFV3 protein, Ndufv3 mRNA, and Ndufv3 exons 1–4 between control and FRAIL1 expressing mouse skeletal muscles. False discovery rate (FDR) was determined by Benjamini‐Hochberg correction. *FDR ≤ 0.05, **FDR ≤ 0.01.

FIGURE 7

Forced FRAIL1 expression downregulates transcripts and proteins involved in the regulation of skeletal muscle structure, energy production, and metabolism. Gene set enrichment analysis (GSEA) of RNA‐seq data was used to identify pathways that were repressed by FRAIL1 expression in mouse skeletal muscle. (a, b) Enrichment plots of the top five Reactome gene sets that were repressed by FRAIL1 expression in mouse tibialis anterior muscles (a) and heatmap of key individual transcripts from those gene sets (b) downregulated in response to FRAIL1. Colored dots left of heatmaps indicate the gene set(s) of origin for the associated gene. (c, d) Enrichment plots of overrepresented cellular component GO terms among skeletal muscle proteins decreased by ectopic FRAIL1 expression (c) and key downregulated proteins from those ontologies (d). False discovery rate (FDR) was determined by Benjamini‐Hochberg correction. *FDR ≤ 0.05, **FDR ≤ 0.01.