INHIBITION OF MICRORNA-92A INCREASES BLOOD VESSELS AND SATELLITE CELLS IN SKELETAL MUSCLE BUT DOES NOT IMPROVE DUCHENNE MUSCULAR DYSTROPHY–RELATED PHENOTYPE IN mdx MICE

Abstract

Introduction:

The vasculature and blood flow in muscle are perturbed in Duchenne muscular dystrophy (DMD) and its mdx mouse model. MicroRNA-92a (miR-92a) is enriched in endothelial cells, especially during ischemic injury.

Methods:

Because antagonizing miR-92a was shown to result in increased proliferation and migration of endothelial cells and recovery from ischemia, we assessed the effects of Antagomir-92a in vitro in muscle stem cell culture and in vivo in mdx mice.

Results:

miR-92a was found to be highly expressed in muscle endothelial cells and satellite cells. Treatment with Antagomir-92a increased capillary density and tissue perfusion, which was accompanied by an increase in satellite cells. However, Antagomir-92a–treated mdx mice showed no histological improvement and had worse muscle function. Antagomir-92a suppressed myogenic differentiation in satellite cell culture.

Discussion:

AntagomiR-92a improves the vasculature but not the muscle in mdx mice, possibly due to its side effects on satellite cell differentiation.

Duchenne muscular dystrophy (DMD) is an X-linked recessive muscle disease caused by a mutation in the gene that codes for dystrophin, a protein that connects the basal lamina to the cytoskeletal proteins. Lack of dystrophin causes contraction-induced injury to the muscle fiber that is repaired by satellite cells. Satellite cells are a stem cell population that are normally quiescent and lie underneath the muscle basal lamina. Upon injury, they proliferate, differentiate into myocytes, and fuse with each other and with the existing muscle fibers to produce regenerating muscle fibers. A small number of these cells return to quiescence in preparation for the next round of injury. The mdx mouse is a genetic model of DMD due a point mutation in the dystrophin gene,^1,2 but displays a much milder dystrophic phenotype, partly due to the robust regenerative capacity of the mouse satellite cells. However, continuous and asymmetrical damage to muscle fibers eventually leads to exhaustion of the satellite cell compartment, resulting in fibrosis and fatty replacement of the muscle tissue.³ Inversely, increasing the number of satellite cells has a beneficial effect on the dystrophic pathology in mdx mice.^4,5 The mdx mouse also has an angiogenic defect leading to a state of functional ischemia.^6–9 Mitigating the angiogenic defect has been shown to improve the dystrophic phenotype.^5,6,10

MicroRNA-92a (miR-92a) is part of the miR-17–92 or Mir17hg-92a pri-miRNA cluster, which is a collection of 5 miRs that are encoded by the MiR17hg cluster in the mouse and transcribed simultaneously. The structure, genetics, and function of the miR-17–92 cluster have been investigated extensively.¹¹ miR-92a is the last encoded miRNA in the cluster, is highly enriched in endothelial cells, and has been implicated in cardiovascular disease. Endothelial cells from patients with coronary artery disease have higher expression of miR-92a, and have an anti-angiogenic function both in vitro and in-vivo.^12,13 As such, miR-92a has been extensively investigated in the context of ischemic disease. Bonauer et al. showed that inhibition of miR-92a using synthetic RNA oligonucleotides against miR-92a (Antagomir-92a) after hindlimb ischemia greatly improved revascularization.^13,14 The vascular niche is important for satellite cell localization and maintenance^15–17; increases in vascular density, blood flow, and satellite cell number have been shown to improve muscle pathology in mdx mice.^5,10,18 However, it remains unknown whether Antagomir-92a treatment can also ameliorate the functional ischemic phenotype seen in the mdx mouse.^8,19 In normal tissue, the effects of miR-92a are thought to be largely endothelial-cell-specific.¹¹ However, their role in skeletal muscle is largely unknown. In this report, we investigated miR-92a in muscle satellite cells and assessed the potential for rescuing the angiogenic defect in the mdx mice after treatment with Antagomir-92a.

METHODS

Bioinformatics.

RNA-seq data from isolated muscle fibers, satellite cells, and endothelial cells were obtained from available data sets (GSE108739). Sequencing reads were mapped using HISAT2 and the alignment and quantification performed using Stringtie.²⁰ RNA-seq data from whole muscles were obtained from previously published data sets (GSE101900).²¹ miRNA sequencing data sets also were obtained from previously published data sets (GSE104284).²²

Mouse Experiments.

B6Ros.Cg-Dmd^mdx-4cv/J (mdx^4cv), B6Ros. Cg-Dmd^mdx-5cv/J(mdx^5cv), and C57BL/6 (wild-type) mice were obtained from Jackson Laboratories. Both mdx^4cv and mdx^5cv mice have approximately 10-fold fewer revertant dystrophin-positive fibers than the original mdx mutant, providing a useful model for investigating a role in DMD angiogenic pathologies.^2,23 Age-matched mdx^4cv male littermates were used for the in-vivo studies. Hindlimb muscles of C57BL/6 and mdx^5cv mice were used to isolate satellite cell–derived primary myoblasts. The animals were housed in a special pathogen-free (SPF) environment and were monitored by Research Animal Resources at the University of Minnesota. All protocols were approved by the institutional animal care and use committee of the University of Minnesota and complied with National Institutes of Health guidelines for the use of animals in research.

Antagomir-92a Administration.

Single-stranded RNA oligonucleotides sequences were commercially synthesized as custom oligonucleotides (Dharmacon) Antagomir-92a (5'-CAGG CCGGGACAAGUGCAAUA-3') and Antagomir-scramble (5'-AA GGCAAGCUGACCCUGAAGUU-3'). All nucleotides were 2' O-methyl modified.¹⁴ The first 2 and last 3 linkages were phosphorothioate and the 3' end had a cholesterol moiety. The Antagomirs were resuspended in sterile phosphate-buffered saline (PBS) and stored at −80°C until use. For in-vivo experiments, mdx^4cv mice were injected with Antagomir-92a (n = 8) or control (scramble-Antagomir) (n = 7) at a dose of 10 mg/kg body weight intraperitoneally every other day, 3 times, starting at postnatal day 12 (p12), as described elsewhere.^12,13 For in-vitro culture experiments, 100 nM of Antagomir-92a or control Antagomir was used and replenished every day.

RNA Extraction and Reverse Transcript Quantitative Polymerase Chain Reaction.

Total RNAs containing miRNA were extracted from wild-type (WT) and mdx^4cv mouse diaphragm muscle and myoblasts after Antagomir-92a administration using the miRNeasy RNA isolation kit (Qiagen). A SYBR Green microRNA expression assay was performed with 500 ng of total RNA using the miScript Reverse Transcription Kit and miScript SYBR Green PCR Kit (Qiagen) according to the manufacturer’s instructions. Specific primers for miRNA expression were purchased from Qiagen. For gene expression analysis, first-strand cDNA was produced using a Transcriptor First Strand cDNA synthesis kit (Roche Molecular Systems), and synthesized cDNA was mixed with GoTaq qPCR Master Mix (Promega). Specific primers for mRNA expression used for PCR have been described previously.^24,25 The expression levels of miRNA and mRNA were quantified on a Realplex 2S system (Eppendorf) following the manufacturer’s instructions. miRNA expression was normalized against the expression of U6, and individual mRNA expression was normalized against the expression of 18S rRNA.

Histology.

Early myofiber membrane permeability leads to fiber damage and subsequent fibrosis in mdx mice.^26–28 Membrane permeability was detected using Evans blue dye (EBD), as described elsewhere.^5,29 EBD (1% PBS solution; Sigma-Aldrich) was injected intraperitoneally 18 h before necropsy. Tibialis anterior (TA) muscle, extensor digitorum longus (EDL) muscle, and diaphragm muscles were dissected and frozen in OCT compound. Eight-micron-thick transverse cryo-sections were used for all analyses. Sirius Red staining was performed as previously described.⁵ Capillary density was detected with the VectaStain Elite ABC Kit (Vector Laboratories) for immunohistochemistry according to the manufacturer’s instructions. Sections were labeled with anti-CD31 (14–0311-85; eBioscience) followed by biotin-conjugated anti-mouse immunoglobulin G secondary antibody. Sections were colorized using 3-amino-9-ethylcarbazole (AEC; Sigma-Aldrich). Immunofluorescence was performed using anti–M-cadherin antibodies (611100; BD Biosciences) followed by anti-mouse Alexa Fluor 488 (Thermo Fisher Scientific). Sections and cell cultures were also nuclear stained with DAPI (Sigma-Aldrich) and mounted with fluorescence mounting media (Dako). Microscopic images were captured by a DP-70 digital camera attached to BX51 fluorescence microscope with 20× and 40× UPlanFLN objectives (all from Olympus). Fiji was used for image processing.³⁰

Laser Doppler Flow.

Red blood cell (RBC) flux was evaluated using the MoorLab laser Doppler flow (LDF) meter with an MP7a probe that allows for collecting light from a deeper tissue level than standard probes, according to the manufacturer’s instructions (Moor Instruments) and as described elsewhere.⁵ Briefly, the fur from the right hindlimb was removed using a chemical depilatory cream. Readings were taken using the probe from at least 10 different spots on the TA muscle. The flux was recorded and data are presented as arbitrary units (AU).

Cell Culture.

Satellite cell–derived primary myoblasts were isolated from adult lower hindlimb muscle from 6-week-old C57BL/6 (WT) and mdx^5cv mice, as described elsewhere.³¹ Cells were maintained on collagen-coated dishes in myoblast growth medium consisting of Ham’s F-10 medium (Sigma-Aldrich) supplemented with 20% fetal bovine serum (FBS; Gibco-Invitrogen) and 20 ng/ml basic fibroblast growth factor (Life Technologies). Proliferation was measured using the Click-IT EDU assay (Thermo Fisher Scientific). Cell cycling was synchronized overnight in Ham’s F-10 media with 0.1% FBS. Cells were then stimulated using myoblast growth medium with 100 nM of Antagomir-92a or scramble-Antagomir control (Dharmacon), or after transfection with miR-92a mimic or miR mimic negative control (Applied Biological Materials) via Poly-Jet reagent (SignaGen Laboratories). 5-ethynyl-2’-deoxyuridine (EDU) 10 μM was added for 8 h. Cells were fixed and assayed for EDU using the manufacturer’s instructions. EDU⁺ cells were counted over 3 microscopic fields for each plate. Each experiment was performed at least in triplicate. To assess differentiation, myoblasts were incubated in 100 nM of Antagomir-92a or scramble-Antagomir control or transfected with miR-92a mimic or miR mimic negative control in growth medium for 1 day, and changed to 5% horse serum (Gibco-Invitrogen) in Dulbec-co’s modified Eagle medium (Gibco-Invitrogen) media for 3 days. Cells were stained using for pan sarcomeric myosin heavy chain (MHC) using the MF20 antibody (Developmental Studies Hybridoma Bank). Fusion index was calculated as MHC⁺ fibers with more than 2 nuclei as a percent of all nuclei.

Grip Strength Test.

Forelimb grip strength test was performed following previously published procedures.³² Briefly, mdx^4cv mice were gently pulled by the tail after forelimb-grasping a metal bar attached to a force transducer (Columbus Instruments). Grip strength tests were performed by the same blinded examiner. Five consecutive grip strength tests were recorded and the mice were returned to their cage for a resting period of 20 min. Then, 3 series of pulls were performed, each followed by a 20-min resting period. The average of the 3 highest values of the 15 values collected was normalized to the body weight for comparison, expressed as Newtons (N) per kilogram.

Statistics.

All data are presented as mean ± SEM. Comparison between groups was done by Student’s t-test using a two-tailed distribution with two-sample equal variance unless otherwise stated. Graphing and statistical analysis was performed using Prism 7 (GraphPad). Data were considered statistically significant at P < 0.05.

RESULTS

Expression of miR-92a in Skeletal Muscle.

We assessed the expression profiles of the miR-17–92a cluster (Mir17hg in mice) in freshly isolated endothelial cells, satellite cells, and muscle fibers from muscle of adult mice. The miR-17–92a cluster was equally expressed in freshly isolated endothelial and satellite cells (Fig. 1A). By comparison, mature muscle fibers expressed much less miR-17–92a cluster. We investigated the expression of miR-92a in whole muscle and found that the miR-17–92a cluster (Fig. 1B) and miR-92a (Fig. 1C) were increased after freeze muscle injury, peaking at 72 h before reaching control levels. By contrast, miR-92a was only increased at the 72-h time-point in isolated satellite cells from injured muscle (Fig. 1D). In fact, miR-92a was the highest expressing miRNA in satellite cells 72 h after injury (Fig. 1E). Taken together, muscle satellite cells express miR-92a in high abundance.

FIGURE 1.

miR-92a expression in skeletal muscle. (A) Expression of miR-17–92a miRNA cluster (Mir17hg) from RNA-seq of freshly isolated endothelial cells, satellite cells, and muscle fibers from WT mice show equal expression of Mir17hg in satellite cells and endothelial cells and significantly lower expression in single fibers. TPM, transcripts per kilobase million. (B) Expression of Mir17hg from RNA-seq of whole muscle after freeze injury to the TA muscle shows transient upregulation of Mir17hg. Expression levels at 10 h and 72 h are significantly different from control with a false discovery rate (FDR) < 0.001. The dotted line represents the value for control. (C) Expression of miR-92a from miRNA-seq in whole muscle after freeze injury to the TA muscle shows a 2-fold increase of miR-92a at 72 h after injury. The dotted line represents the value for control. (D) Expression of miR-92a from miRNA-seq in isolated satellite cells after freeze injury to the TA muscle shows over 10-fold increase of miR-92a at 72 h after injury. The dotted line represents the value at 3 h postinjury. (E) Heat map shows top 16 miRNAs expressed by satellite cells 72 h postinjury. miR-92a is the highest expression of microRNA in satellite cells. *P < 0.05 and ***P < 0.001.

Inhibition of miR-92a Improves Vascularization in mdx^4cv Mice.

We examined blood flow in age-matched littermate control mdx^4cv mice and mice in which we delivered Antagomir-92a or scramble-Antagomir as a control into the intraperitoneal cavity 3 times at ages p12–p16 (Fig. 2A). We validated the miR-92a expression in diaphragm muscle after Antagomir-92a administration by reverse transcript quantitative polymerase chain reaction (RT-qPCR), which showed increased miR-92a expression in diaphragm muscle of mdx^4cv mice compared with WT mice (Fig. 2B). Importantly, RT-qPCR showed a significant reduction of miR-92a expression after Antagomir-92a administration in mdx^4cv. We also performed RT-qPCR for expression of DKK3 and KLF4 as known miR-92a target genes,^33,34 both which were slightly upregulated after Antagomir-92a administration (Fig. 2C).

FIGURE 2.

Antagomir-92a treatment increases angiogenesis in mdx^4cv mice. (A) Schematic time course for Antagomir-92a administration in mdx^4cv mice. (B) Comparison of expression of basal levels of miR-92a in WT vs. mdx^4cv diaphragm muscle by RT-qPCR (n = 3 in each group). (C) Validation of expression of miR-92a, DKK3, and KLF4 in mdx^4cv diaphragm muscle by RT-qPCR after Antagomir-92a administration (n = 3 each group). (D) Representative photomicrographs of diaphragm muscle sections immunostained with an antibody to CD31 in the muscle of control mdx^4cv and Antagomir-92a–treated mdx4cv mice. Scale bar = 50 μm. (E) Quantification of CD31⁺ cells in cross-sections shows Antagomir-92a–treated mice (n = 4) have higher capillary density compared with control (n = 6). (F) RBC flux from laser Doppler flow shows significant increased perfusion in the TA muscle of mdx4cv mice (n = 6 in each group). *P < 0.05 and **P < 0.01.

We performed skeletal muscle LDF imaging at p35 and before necropsy, and measured forelimb grip strength at p45, the time of onset of dystrophic changes in the mdx^4cv mice (Fig. 2A). We found that administration of Antagomir-92a could also increase the capillary density in the mdx^4cv skeletal muscle at its basal state during postnatal muscle development. We quantified the number of capillaries, as determined by CD31 immunostaining on cryostat sections of diaphragm muscle from mdx^4cv mice after treatment with Antagomir-92a compared with untreated mdx^4cv control mice. Antagomir-92a–treated muscle had significantly higher capillary density compared with control (538.6 ± 30.05 vs. 406.3 ± 22.15; Fig. 2D and E). In addition, Antagomir-92a treatment resulted in a physiological increase in perfusion as measured by laser Doppler RBC flux in the TA muscle at 70.31 ± 8.742 units compared with 55.32 ± 2.95 units in the control mdx^4cv mice, a 27.1% difference (Fig. 2F). Therefore, inhibition of miR-92a treatment resulted in increased capillary density and increased vascular perfusion in the mdx^4cv mouse muscle.

Inhibition of miR-92a Increases Satellite Cells In Vivo and Decreases Myogenic Differentiation In Vitro.

We examined the effects of increased vasculature in the mdx^4cv mouse skeletal muscles on the resident satellite cells. We performed immunofluorescence to detect satellite cells on cross-sections of the TA muscle. We found an increase in the number of M-cadherin-positive satellite cells present in the TA muscle in the mdx^4cv mice after treatment with Antagomir-92a compared with the control mdx^4cv mice (Fig. 3A and B). To elucidate whether this was due to a direct effect of Antagomir-92a specifically on the satellite cells, we isolated satellite cells from both WT and mdx^5cv mice and cultured them as primary myoblasts in vitro. We treated the cells with Antagomir-92a, and validated miR-92a expression in cells by RT-qPCR. The data showed significant reduction of miR-92a expression after Antagomir-92a treatment in both WT and mdx^5cv myoblasts (Fig. 3C). We also performed RT-qPCR for expression of DKK3 and KLF4 as miR-92a target genes, and showed increased expression of both genes after Antagomir-92a treatment.

FIGURE 3.

Antagomir-92a increases satellite cell number in vivo and decreases myogenic differentiation in vitro. (A) Representative photomicrographs showing the satellite cell marker M-cadherin in TA muscle cross-sections. Arrows denote M-cadherin⁺/DAPI⁺ satellite cells. Scale bar = 10 μm. (B) Quantification of satellite cells in mdx^4cv mice after treatment with Antagomir-92a shows an increase in satellite cell numbers in mdx^4cv mice (n = 5 in each group). (C) Validation of expression of miR-92a, DKK3, and KLF4 in WT and mdx^5cv by RT-qPCR after Antagomir-92a administration (n = 3 in each group). (D) Representative figures for proliferation and differentiation in vitro. (Top) Myoblasts stained for EDU (green) and DAPI (blue) in the mdx^5cv myoblasts show no difference in proliferation after Antagomir-92a or miR-92a mimic administration. (Bottom) Under the differentiation conditions, mdx^5cv myoblasts show reduced MHC⁺ differentiation after Antagomir-92a administration, but enhanced differentiation after transfection with miR-92a mimic. Scale bars = 100 μm. *P < 0.05 and **P < 0.01.

We determined proliferation rate in the cultured satellite cells using EDU incorporation after Antagomir-92a treatment. Quantitative analysis of these labeled cultures showed there were no significant differences in the proliferation rate of myoblasts compared with the untreated control cells in either the WT or the mdx^5cv myoblasts (Figs. 3D and 4A and B). Next, we assessed the influence of Antagomir-92a on myogenic differentiation. We found that inhibiting miR-92a significantly inhibited myogenic differentiation of the myoblasts from both WT and mdx^5cv mice (Figs. 3D and 4C). This also led to a decrease in fusion of the differentiated myotubes in both WT and mdx^5cv myotubes (Figs. 3D and 4E). By contrast, miR-92a mimic transfections into WT myoblasts showed significantly enhanced myogenic differentiation and myoblast fusion (Figs. 3D and 4D and F), indicating the promyogenic differentiation function for miR-92a. Therefore, treatment with Antagomir-92a resulted in an increased number of satellite cells in vivo and inhibition of myogenic differentiation in vitro.

FIGURE 4.

Myogenic differentiation is suppressed by Antagomir-92a but promoted by miR-92a mimic in vitro. (A) Quantification of EDU⁺ proliferating myoblast in culture shows no effect of inhibiting miR-92a in WT (n = 3)or mdx^5cv (n = 4) myoblasts. (B) Quantification of EDU⁺ proliferating myoblasts in culture shows no effect of mimic of miR-92a in WT (n = 3) or mdx^5cv (n = 3) myoblasts. (C) Quantification of MHC⁺ differentiated myoblast in cultures shows a decreased differentiation by inhibition of miR-92a in WT (n = 4) and mdx^5cv (n = 4) myoblasts. (D) Quantification of MHC⁺ differentiated myoblast in cultures shows enhanced differentiation after transfection with miR-92a mimic in WT (n = 3) or mdx^5cv (n = 3) myoblasts. (E) Quantification of fusion index in differentiating myoblasts in culture show decreased fusion by inhibition of miR-92a in WT (n = 3) or mdx^5cv (n = 5) myoblasts. (F) Quantification of fusion index in differentiating myoblasts in culture shows enhanced differentiation after transfection with miR-92a mimic in WT (n = 3) or mdx^5cv (n = 3) myoblasts. *P < 0.05 and **P < 0.01.

Inhibition of miR-92a Does Not Improve Dystrophic Phenotype in mdx^4cv Muscle.

We next assessed muscle pathology in the Antagomir-92a–treated mdx^4cv mice. We did not find any significant differences in body weight (Fig. 5A) or muscle weight (data not shown) in the Antagomir-92a–treated mdx^4cv mice compared with the age-matched control mdx^4cv mice. However, we found signs of corneal cataracts (data not shown). We used grip strength as a measure of muscle function in the mdx^4cv mice treated with Antagomir-92a compared with the mdx^4cv controls and found that the treated mdx^4cv mice exhibited decreased forelimb grip strength, which was normalized to body weight (Fig. 5B).

FIGURE 5.

Antagomir-92a treatment does not improve dystrophin pathology in the mdx^4cv mice. (A) Body weight measurements for mdx^4cv mice during treatment with AntagomirR-92a show no difference in body weight compared with mdx^4cv controls (n = 4 in each group). (B) Antagomir-92a–treated mdx^4cv mice show decreased forelimb grip strength at 6 weeks of age compared with control mdx^4cv mice (n = 4 in each group). (C) Representative images of diaphragm muscle from mdx^4cv mice treated with Antagomir-92a show membrane permeability (EBD, top) and fibrosis (Sirius Red, bottom). Scale bars = 200 μm. (D) Quantification of EBD⁺ fibers in diaphragm muscle cross-sections of Antagomir-92a–treated mdx^4cv mice (n = 4) show no improvement in muscle membrane stability compared with control mdx^4cv mice (n = 6). Scale bars = 200 μm. (E) Quantification of fibrotic areas in diaphragm muscle cross-sections of Antagomir-92a–treated mdx^4cv mice show no improvement compared to control mdx^4cv mouse muscles (n = 7 each group). *P < 0.05 and **P < 0.01.

We also performed histological analysis of the treated and control mdx^4cv muscles to investigate the potential for Antagomir-92a to alter muscle pathology in the mdx^4cv mouse. We found no difference in the EBD-positive area in the Antagomir-92a–treated muscle vs. control, at 6.935 ± 1.646 compared with 7.501 ± 3.09, respectively (Fig. 5C and D). We quantified the fibrotic changes and found no significant difference in the total fibrotic area in Antagomir-92a–treated mdx^4cv muscle vs. control mdx^4cv muscle, at 1.876 ± 0.2808 compared with 1.926 ± 0.1795 (Fig. 5C and E).

DISCUSSION

The mdx mouse model has been shown to have a vascular defect, which contributes to the DMD muscle pathology.^5–10 The miRNA, miR-92a, was shown to be enriched in endothelial cells and its inhibition led to rescue from induced ischemic injury.^12,13 In this report, we have shown that miR-92a is expressed in muscle satellite cells in amounts comparable to endothelial cells and that its expression is transiently increased during muscle injury in whole muscle and satellite cells. Inhibition of miR-92a has previously been shown to increase revascularization after ischemic injury in both small and large animals.^13,35,36 We showed that blocking miR-92a using Antagomir-92a increases the vascularity in dystrophic skeletal muscle. This increase in vascularity is accompanied by an increase in satellite cell number. The latter does not seem to be a direct effect of Antagomir-92a on satellite cell proliferation, as we recently reported that increased juxavascular niche enhances satellite cell self-renewal.¹⁷ Thus, the in-vivo increase in satellite cells may be due to costimulation from endothelial cells or inhibition of myogenic differentiation. Last, Antagomir-92a did not improve early or late histological signs of muscle pathology and, through a mechanism not fully understood, resulted in decreased grip strength in the treated mdx mice compared with controls.

Although miR-92a has been well studied in the context of cardiovascular disease and induced ischemic injury, its effect on skeletal muscle is less understood. Sengul et al. investigated the effect of antagomir-92a on skeletal muscle and found no difference in capillary density in the Antagomir-92a–treated mice, using the same sequence as we did, despite seeing a reduction in tissue miR-92a levels.³⁷ One possible reason for the differences between these 2 studies is that we used a higher dose of Antagomir-92a, 10 mg/kg body weight, as compared with 2.7 mg/kg body weight. Even though treatment with Antagomir-92a in neonatal mice resulted in a significant increase in vasculature, satellite cell number, and blood flow, the dystrophic phenotype in the mdx mice could not be improved by Antagomir-92a treatment. We showed that treatment with Antagomir-92a inhibited myogenic differentiation of satellite cells in vitro. In addition, treatment with miR-92a mimic showed that miR-92a functions as a pro-myogenic differentiation microRNA. One possibility for the lack of amelioration of the dystrophic pathology is that reduced myogenic differentiation after Antagomir-92a treatment may affect the development and maintenance of the dystrophic phenotype in mdx mice. A second possibility is that the skeletal muscles in the mdx mouse also have an inflammatory component, and this may be worsened by treatment with Antagomir-92a.^36,38 The third possibility is that increased vascularity and blood flow in the normal situation may not be enough for an efficient therapeutic strategy for DMD. Because mdx mice undergo muscle ischemia after muscle contraction,^8,9 functional improvement of vasculature, including increased blood flow in ischemic conditions, is crucial, in which Antagomir-92a treatment can be combined with phosphodiesterase-5 (PDE5) inhibitor therapy.⁸ Other studies also demonstrated the oncogenic effects of miR-92a, including promotion of cancer growth and inhibition of apoptosis via suppression of p21, KLF2, KLF4, p63, DKK3, PTEN, FOXP1, RBM4, and BCL2L11 genes.^{33,34,39–44} In particular, DKK3 has been identified as a diagnostic marker and as a therapeutic target for age-related muscle atrophy,²⁵ suggesting that upregulation of DKK3 via Antagomir-92a administration may induce worsened dystrophic phenotype in mdx mice.

Our study highlights that skeletal muscle-specific effects of 1 need to be more carefully considered. Recruitment for a phase 1 clinical trial of miR-92a inhibition for heart failure started in 2018.⁴⁵ Despite the increase in capillary density and satellite cell number, our current study has failed to show any improvement of dystrophic phenotype in mdx mice after treatment with Antagomir-92a. This is a major consideration when miR-92a is used as a systemic treatment as potential negative sequelae in the skeletal muscles maybe an unintended consequence, and thus should be tracked as part of the study design.

Abbreviations:

AEC

3-amino-9-ethylcarbazole

DMD

Duchenne muscular dystrophy

EDU

5-ethynyl-2’-deoxyuridine

EBD

Evans blue dye

EDL

extensor digitorum longus

FBS

fetal bovine serum

FDR

false discovery rate

LDF

laser Doppler flow

MHC

myosin heavy chain

Mir17hg

MiR-17–92a cluster

miR-92a

microRNA-92a

p12

postnatal day 12

PBS

phosphate-buffered saline

PDE5

phosphodiesterase-5

RBC

red blood cell

RT-qPCR

reverse transcript quantitative polymerase chain reaction

SPF

special pathogen-free

TA

tibialis anterior

WT

wild-type