Targeting TGF-β Signaling by Antisense Oligonucleotide-mediated Knockdown of TGF-β Type I Receptor

Abstract

Duchenne muscular dystrophy (DMD) is caused by lack of functional dystrophin and results in progressive myofiber damage and degeneration. In addition, impaired muscle regeneration and fibrosis contribute to the progressive pathology of DMD. Importantly, transforming growth factor-β (TGF-β) is implicated in DMD pathology and is known to stimulate fibrosis and inhibit muscle regeneration. In this study, we present a new strategy to target TGF-β signaling cascades by specifically inhibiting the expression of TGF-β type I receptor TGFBR1 (ALK5). Antisense oligonucleotides (AONs) were designed to specifically induce exon skipping of mouse ALK5 transcripts. AON-induced exon skipping of ALK5 resulted in specific downregulation of full-length receptor transcripts in vitro in different cell types, repression of TGF-β activity, and enhanced C2C12 myoblast differentiation. To determine the effect of these AONs in dystrophic muscles, we performed intramuscular injections of ALK5 AONs in mdx mice, which resulted in a decrease in expression of fibrosis-related genes and upregulation of Myog expression compared to control AON-injected muscles. In summary, our study presents a novel method to target TGF-β signaling cascades with potential beneficial effects for DMD.

Figure 1

Schematic overview of transforming growth factor-β (TGF-β) signaling cascades and the effect of the designed antisense oligonucleotides (AONs). (a) Overview of the mechanism of TGF-β signaling. TGF-β, MSTN, and activin bind to type II and type I receptors, which are activated and induce downstream Smad2/3-dependent signaling pathways. Yellow stars depict phosphorylation. (b) Overview of ALK5 transcript and the corresponding protein domains, showing the effect of AON-mediated exon skipping. The ALK5 gene consist of nine exons. The different corresponding protein domains are shown below the transcript. Receptor domain (RD, in green), glycine/serine-rich domain (GS, in red) and kinase domain (kinase, in blue). AON-mediated targeting of exon 2 results in transcripts that lack the sequence encoding the ligand-binding domain. AON-mediated targeting of exon 6 disrupts the open reading frame and generates a premature stop triplet (STOP).

Figure 2

In vitro proof of principle of antisense oligonucleotides (AON)-mediated ALK5 exon skipping. (a) AONs targeting exon 2 of ALK5 (ALK5 AON) were transfected at different concentrations into C2C12 mouse myoblasts. Two days after transfection, the efficacy to induce exon skipping was assessed by reverse transcription PCR ALK5-specific primers in flanking exons (red arrowheads). Nontransfected (NT) cells or cells transfected with control AON (Ctrl AON, nontargeting) served as controls. (b) Sequencing of excised PCR products showed exclusion of the targeted exons ALK5 AON transfection (exon 2 skip) in C2C12 cells. (c) Quantitative real-time PCR of C2C12 samples was performed to compare full-length ALK5 transcript expression using primers in the skipped exon and in the flanking exon. Cells were transfected with 200 nmol/l AON. These data are shown as the average of at least three independent experiments and is shown relative to control AON samples. Error bars represent standard deviations. *P < 0.05; **P < 0.01

Figure 3

ALK5 antisense oligonucleotides (AONs) inhibit MSTN and transforming growth factor-β (TGF-β) signaling. (a) ALK5 AONs (200 nmol/l) specifically repressed TGF-β-induced (CAGA)₁₂-luciferase activity in both C2C12 myoblasts and C3H10T1/2 cells. (b) MSTN-induced Smad3-dependent (CAGA)₁₂-luciferase activity was specifically repressed by ALK5 AONs (200 nmol/l) in C3H10T1/2 mesenchymal stem cells. In contrast, ALK5 AONs did not inhibit MSTN activity in C2C12 myoblasts. Firefly luciferase activity of (CAGA)₁₂-luciferase constructs was corrected with renilla-luciferase activity of cotransfected cytomegalovirus-renilla constructs. The experiments were performed at least three times and each experiment was performed in triplicate. The values were averaged and plotted as fold relative to control AON (Ctrl AON)-transfected samples. Error bars represent standard deviations. *P < 0.05, **P < 0.01.

Figure 4

ALK5 antisense oligonucleotides (AONs) enhance myoblast differentiation. (a) Immunofluorescence images of C2C12 cells transfected with control (Ctrl) or ALK5 AONs at different time points after initiation of myogenic differentiation. Immunofluorescent staining is shown for desmin (myogenic cells, red), myosin (differentiated myotubes, green), and 4',6-diamidino-2-phenylindole (nuclear, blue). (b) Fusion and differentiation indices were calculated based on the immunofluorescence images. ALK5 AONs increase fusion and differentiation index in C2C12 cells. Error bars represent standard deviations. **P < 0.01; ns, not significant

Figure 5

Effect of ALK5 antisense oligonucleotides (AONs) after intramuscular injections in mdx mice. (a) Quantitative polymerase chain reaction analysis of full-length ALK5 expression in samples from ALK5 or scrambled control AON (ScrALK5 AON)-injected triceps muscles. Muscles were isolated and analyzed 10 days after the last injection. Specific reduction of full-length ALK5 was observed after injection with ALK5 AONs. (b) Treatment with ALK5 AONs increased the expression of Myog (not significant) and decreased expression of Col1a1. ALK5 AONs reduced expression levels of Ctgf (not significant) and Serpine1. (c) Protein lysates were analyzed for Serpine1 and phosphorylated Smad2 by western blot and quantified densitometrically using actin for normalization. Error bars in a, b, and d represent standard deviations. *P < 0.05; **P < 0.01.