Targeting microRNA-mediated gene repression limits adipogenic conversion of skeletal muscle mesenchymal stromal cells

Abstract

Intramuscular fatty deposits, which are seen in muscular dystrophies and with aging, negatively affect muscle function. The cells of origin of adipocytes constituting these fatty deposits are mesenchymal stromal cells, fibroadipogenic progenitors (FAPs). We uncover a molecular fate switch, involving miR-206 and the transcription factor Runx1, that controls FAP differentiation to adipocytes. Mice deficient in miR-206 exhibit increased adipogenesis following muscle injury. Adipogenic differentiation of FAPs is abrogated by miR-206 mimics. Using a labeled microRNA (miRNA) pull-down and sequencing (LAMP-seq), we identified Runx1 as a miR-206 target, with miR-206 repressing Runx1 translation. In the absence of miR-206 in FAPs, Runx1 occupancy near transcriptional start sites of adipogenic genes and expression of these genes increase. We demonstrate that miR-206 mimicry in vivo limits intramuscular fatty infiltration. Our results provide insight into the underlying molecular mechanisms of FAP fate determination and formation of harmful fatty deposits in skeletal muscle.

Keywords: adipogenesis; aging; cell fate determination; fatty infiltration; fibroadipogenic progenitor; mesenchymal stromal cell; muscular dystrophy; skeletal muscle; stem cell.

Figure 1.. miR-206 as a putative regulator of the adipogenic fate transition of FAPs to adipocytes

(A) Representative FACS plot of the gating for the isolation of FAPs. (B) Quantification of Perilipin+ cells when grown in basal medium (BM) or adipogenic differentiation medium (ADM) for 6 days (n = 4). (C) miRNAs detected in miRNA-seq when comparing FAPs in ADM or BM. Red lines denote the 5-fold linear threshold (2.32 Log₂). (D) Heat map showing miRNAs that have a ≥ 5-fold differential expression in FAPs grown in ADM relative to those in BM. (E) Quantification of miR-206 expression by qRT-PCR in freshly isolated (FI) cells, and cells grown in BM or ADM for 24 hours (n = 3 per condition). (F) Representative images of cells cultured in BM with inhibitors or ADM with mimics, and assayed for the adipogenic marker, PPARγ (nuclear localized bright green dots). Insets are magnified views of areas noted by the dotted boxes. Quantification of replicates are shown in the graphs (n = 6 for each condition). Scale bars are 100 μm and 50 μm for insets. Error bars represent +/− SEM. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001.

Figure 2.. Increased intramuscular fatty infiltration following injury is observed in miR-206 KO mice

(A) Tibialis anterior (TA) cross sections stained with Perilipin to identify adipogenic tissue (dpi – days post injury). The Perilipin-stained area for each replicate is quantified in the graph (n = 3–9). (B) Images of WT and miR-206 KO FAPS transfected with miR-206 mimics or control (Ctrl) mimics and cultured in BM for 4 days. Fixed cultures were stained with PPARγ and Perilipin to detect adipogenic cells and the relative percent marker+ cells for replicates are presented in the graphs (n = 5). Scale bars (A) are 500 μm and (D) are 100 μm. Error bars represent +/− SEM. ns, non-significant; * p ≤ 0.05, ** p ≤ 0.01.

Figure 3.. Modulating miR-206 in transplanted FAPs limits their adipogenic conversion in vivo.

(A) Quantification of miR-206 relative expression by qRT-PCR in freshly isolated cells from muscle prior to injury (uninjured (UI)) and at the indicated time after injury (n = 4–6 for each time point). (B) Schematic outlining the isolation of GFP+ cells, transfection with miRNA mimics, and transplantation of these cells into glycerol injured TAs. (C) Images of fatty areas in sections containing transplanted GFP+ cells transfected with the noted miRNA mimic. Magnified view is of the dotted boxed area and yellow arrow heads note GFP+ Perilipin+ adipocytes. (D) Quantification of transplant replicates (n = 6 for each mimic). Scale bars are 100 μm. Error bars represent +/− SEM. * p ≤ 0.05, *** p ≤ 0.001.

Figure 4.. FAPs without miR-206 are primed for the adipogenic program and Runx1 is a putative target of miR-206

(A) Differential expression of transcripts from RNA sequencing when comparing miR-206 KO versus WT cells. Point encircled with blue notes Ifi202b. (B) The top ten Ingenuity Pathway Analysis (IPA) Canonical Pathways are shown using the enriched transcription factors from the RNA-seq data in (A). (C) Diagram of the LAMP-seq protocol. (D) Transcript enrichment when comparing results from the miR-206 versus miR-Control (Ctrl) pulldown. (E) Heat map illustrates the top 20 transcription factors (TFs) enriched in PD. Ctrl, control pulldown; PD, miR-206 pulldown. (F) Sequence alignment between the mature miR-206 sequence and the Runx1 transcript (nucleotides 462–487) and the area of predicted interaction (vertical lines between complementary base pairs). (G) The graph displays results from the miR-206-transcript interaction assay, where MMD, Crispld2, Runx1, or Runx1-mutant transcripts were assayed (n = 4–8). Error bars represent +/− SEM. ** p ≤ 0.01, **** p ≤ 0.0001.

Figure 5.. Runx1 translation is inhibited by miR-206 and regulates FAP adipogenesis

(A) Runx1 transcript expression measured by qRT-PCR in miR-206 KO cells transfected with control mimics (Ctrl) or miR-206 mimics (n = 6 per condition). (B) Runx1 protein levels in miR-206 KO cells transfected with control or miR-206 mimics. (n = 3 with cells pooled from 3 mice). (C) Runx1 transcript levels in miR-206 KO FAPs measured by qRT-PCR (n = 3 per condition). (D) PPARγ+ and Perilipin+ adipogenic cells in miR-206 KO FAPs transfected with control or Runx1 siRNAs. Percent marker+ adipogenic cells for replicates are shown in the graph (n = 4). (E-H) Runx1 overexpression (OE) in WT FAPs. (E and F) Protein levels from cells transduced with control or Runx1 (n = 3 with cells pooled from 3 mice). (G) WT FAPs transduced with control or Runx1 OE virus and stained with Oil Red O (ORO) for adipocytes. The magnified views are from the areas noted by dotted boxes. (H) Quantification of ORO+ cells for each replicate (n = 4 with cells pooled from 3 mice). Scale bars are 100 μm. Error bars represent +/− SEM. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001.

Figure 6.. In vivo miR-206 mimicry limits fatty infiltration in skeletal muscle

(A) TA cross sections from TAs injected with glycerol and with miR-206 or control mimics and stained for perilipin to detect fatty infiltration. Insets are magnified views of the boxed areas. The bracket in the top left panel notes an example of necrotic fibers that absorb antibody. (B, C) The myofiber cross sectional areas (CSAs) were quantified for each replicate and shown as (B) the mean CSA of all myofibers or (C) discrete CSAs of myofibers grouped by size (n = 5). (D) Quantification of the Perilipin+ area for each replicate (n = 5). Ctrl – control. Scale bars are 200 μm and 100 μm for insets. Error bars represent +/− SEM. *** p ≤ 0.001