Background
Duchenne muscular dystrophy (DMD) is a progressive and fatal X-linked recessive disorder of skeletal and cardiac muscles. It is a particularly severe and common form of muscular dystrophy, affecting one in 3500 males at birth [
1]. Mutations in the gene encoding the dystrophin lead to a lack of this protein, which normally ensures the essential link between the subsarcolemmal cytoskeleton and the extracellular matrix at the muscle fibre membrane [
2,
3]. DMD is characterized by repeated cycles of necrosis/regeneration of muscle fibres, progressive replacement of skeletal muscle by fibrotic and adipose tissues and generalized muscle weakness, paralysis and death [
4]. Recently, several gene and cell-based strategies have been developed to restore dystrophin expression in the Golden Retriever muscular dystrophy (GRMD) dog, the clinically relevant animal model of DMD. Some of these innovative approaches have now entered preclinical studies [
5,
6]. In parallel, numerous studies are ongoing to define muscle molecular signatures that could be used to characterize dystrophic dog tissue [
7,
8] or to validate the effect of promising therapeutic strategies [
9,
10].
MicroRNAs (miRNAs) are short non-coding RNA sequences of 21 to 25 nucleotides that regulate gene expression at a post-transcriptional level. Through binding to target mRNA, they promote its degradation or translational inhibition [
11,
12]. In muscle, specific miRNAs (known as myomiRs), such as miR-1, miR-133 and miR-206, are involved in regulation of the proliferation or differentiation of myogenic cells [
13‐
16] and are especially regulated by transcription factors implicated in muscle growth and development [
17,
18]. Other miRNAs, such as miR-29, miR-34, miR-222 and miR-486, also play key-roles in modulating important pathways of skeletal muscle processes [
19‐
22]. Over the last few years, miRNAs have been found to be deregulated in muscular dystrophies [
23,
24]. A specific DMD signature has been identified based on eleven miRNAs that are deregulated both in
mdx mice and DMD patients [
22]. As regards myomiRs, several studies report that miR-1 and miR-133 are under-expressed, while miR-206 is over-expressed in
mdx muscles [
25‐
27]. All these findings indicate an important role of miRNAs in pathophysiological pathways regulating muscle response to damage and regeneration. However, except for a preliminary study performed on
CXMDJ dog muscle [
26], there is currently no experimental data concerning miRNA status specifically in GRMD dog skeletal muscle. Alternatively, a recent study identified deregulated miRNAs in the serum of GRMD dogs [
28]. Although GRMD dogs more closely mimic the human disease than
mdx mouse, the lack of data on this large animal model represents a real limitation for the accurate description of the dysregulation of miRNAs in a DMD-like context. Moreover, it is important to fill this gap in our knowledge of the GRMD dog model, in particular with regard to the preclinical evaluation of new therapeutic proposals.
We show that systemic delivery of MuStem cells (which are muscle-resident stem cells isolated from healthy dog based on delayed adhesion properties) represents an attractive avenue for future therapeutic applications in DMD patients. Indeed, allogeneic MuStem cell transplantation in GRMD dogs leads to reduced muscle damage, increased regeneration activity, and a persistent stabilization of clinical status [
29]. In a previous study, we revealed the impact of MuStem cell transplantation, with an up-regulation of genes reflecting a sustained enhancement of muscle regeneration [
30]. In addition, MuStem cells can act on several other biological pathways implicated in protein degradation mechanisms and energy metabolism, evoking a diffuse impact with a large number of targeted biological processes.
In the present study, we firstly aim at defining the miRNA pattern in the skeletal muscle of 9-month-old GRMD dogs corresponding to an advanced state of the disease. Secondly, we attempt to determine how this pattern could be affected following the intravenous delivery of MuStem cells. We determine, for the first time, that miR-222 displays a differential expression pattern in GRMD dog muscle as shown by its marked up-regulation. Using in situ hybridization, we show that miR-206 and miR-486 are mainly expressed in clusters of newly regenerated fibres. In addition, we demonstrate an up-regulation of both miR-133 and miR-222 4 months after MuStem cell transplantation, highlighting their potential use as novel markers for the follow-up of effects associated with MuStem cell delivery in a dystrophic context.
Methods
Ethics statement and animals
This study was approved by the Ethics Committee on Animal Experimentation of the Pays de la Loire Region (France) in accordance with the guidelines from the French National Research Council for the Care and Use of Laboratory Animals (Permit Number: CEEA.2012.104). All the dogs were obtained from the Centre d'élevage du Domaine des Souches (Mézilles, France), which were kept at the Boisbonne Centre for gene and cell therapy of Oniris (Nantes, France). Fourteen 9-month-old golden retriever dogs were included in the study; five were healthy and nine were GRMD. Affected dogs were identified in the 1st week of life using polymerase chain reaction (PCR)-based genotyping. This identification was corroborated by a dramatic and early rise in levels of serum creatine kinase [
31]. GRMD dogs were divided into three subsets: three GRMD dogs received neither an immunosuppressive regimen nor cell transplantation (subset denoted as GRMD), three received only a continuous immunosuppressive regimen (mock GRMD) and the remaining three received MuStem cell transplantation under immunosuppression (GRMD
MuStem) (Table
1).
Table 1
Description of the fourteen male dogs included in the miRNA study
Physiopathology | Healthy | 5 | healthy | None | None |
GRMD | 3 | dystrophic | None | None |
MuStem cell impact | mock GRMD | 3 | dystrophic | Yes | None |
GRMDMuStem | 3 | dystrophic | Yes | Yes |
Cell delivery procedure
MuStem cells were isolated from a pool of hindlimb muscles of 2.5-month-old healthy dogs and prepared as previously described [
29]. Three GRMD dogs (ranging from 3.5 to 4.5 months of age) were submitted to a continuous cyclosporine-based immunosuppressive regimen as established by Rouger et al. (2011) and received three cell injections (spaced at an interval of 2 weeks): GRMD
MuStem. Each of these injections delivered 5.5×10
7 to 8.0×10
8 MuStem cells/kg into the cephalic vein using laminar flow at a rate of 12 mL/min.
Clinical follow-up
A clinical score was measured weekly for all GRMD dogs following a previously described method [
29,
32]. Dogs were weighted and clinically assessed in a non-blinded manner by a veterinarian on a weekly basis during all the experiment. Briefly, this clinical score is based on 11 locomotion and muscular criteria and 6 items related to general health status. It is expressed as a percentage of the maximum score defined as 100 % for a healthy dog.
Muscle sampling
Biceps femoris muscle samples (0.5 cm
3 fragments) were collected surgically from the middle portion of the muscle in 9-month-old (37 ± 5-week-old) healthy, GRMD, mock GRMD and GRMD
MuStem dogs. The
Biceps femoris is a large and easily accessible muscle. This time-point corresponds to 4 months after systemic administration to the GRMD
MuStem dogs and is the same as in our previous transcriptomic study [
30]. Muscle fragments were divided into two parts for histological and molecular analyses, and subsequently stored at −80 °C until processing.
Histological analysis
Eight μm-thick cryosections were incubated with mouse primary antibody directed against the developmental myosin heavy chain (MyHCd, 1/20, Novocastra, Newcastle, United Kingdom) for 1 h at 37 °C. After successive incubation with a secondary biotinylated antibody and streptavidin horseradish peroxidase conjugate (1/300, Dako, Glostrup, Denmark), MyHCd protein was visualized by diamidinobenzidine tetrahydrochloride (DAB; Dako). Slides were then dehydrated and mounted in a dry mounting medium. Morphometric analysis was performed using a digital camera (Nikon DXM 1200, Nikon Instruments, Badhoevedorp, The Netherlands) combined with image-analysis software (NIS, Nikon). Microscopic fields were randomly selected on immunolabelled sections using intermediate magnification to observe at least 100 fibres. To determine the percentage of MyHCd
+ fibres, at least 662 fibres (1030 ± 125) were counted on three randomly selected microscopic fields. For each measurement, reproducibility is better than 92 %. For dystrophin labelling, all acquisitions were performed with the same signal amplification resulting from identical detector gain, as previously described [
29]. To determine the proportion of dystrophin
+ fibres, 880 ± 101 total fibres were counted (laminin red fluorescent immunolabelling) in the
Biceps femoris muscle sections of the GRMD
MuStem dogs (
n = 3) and then the number of fibres expressing dystrophin was determined from DYS2 (Novocastra) green fluorescent immunolabelling.
miRNAs isolation and qPCR
miRNAs were extracted from muscle samples of right and left Biceps femoris of each animal. The mirVana miRNA isolation kit (Ambion, Austin, TX, USA) was used, according to the manufacturer’s instructions, and microRNAs were finally eluted with 100 μL of water and quantified using a Nanodrop spectrophotometer (Labtech, Wilmington Delaware, USA). Reverse Transcription reactions were carried out on 5 ng of miRNAs using the TaqMan miRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) and miRNA-specific stem loop primers for miR-1, miR-133a, miR-206, miR-222, and miR-486 (Applied Biosystems miRNA assays). Real-time PCR reactions were performed at least in duplicate with miRNA-specific primers and Taqman® probes on the CFX96 PCR System (BioRad). Data were normalized using U6 snRNA (RNU6B) as an internal control and differential expression was calculated using the 2-∆∆Ct method. For each miRNA, statistical differences between two groups were analysed by a Mann–Whitney test.
In situ hybridization
In situ hybridization (ISH) was performed on muscles from healthy, GRMD, mock GRMD and GRMDMuStem dogs, using digoxigenin (DIG)-labelled miRCURY locked nucleic acid (LNA) detection probes (Exiqon, Vedbaek, Denmark), corresponding to hsa-miR-486 (38596–05), hsa-miR-206 (88081–15) and scramble-miR (99004–05 and 99004–15). Ten μm-thick frozen muscle sections were air-dried for 1 h and fixed for 10 min in 4 % paraformaldehyde. Then, the sections were permeabilized with proteinase K (20 μg/mL) for 10 min. For pre-hybridization, the tissue sections were covered for 1 h with hybridization buffer containing 50 % formamide, 4X SSC, 1X Denhardt’s solution, 500 μg/mL salmon sperm DNA (Sigma-Aldrich, Saint Quentin Fallavier, France), 10 % dextran sulfate and 1X Blocking Reagent (Roche, Basel, Switzerland). For hybridization, 50 nM of DIG-labelled probes diluted in hybridization buffer were applied per section and incubated in a sealed humidified chamber for 16 h at 55 °C. A stringency wash was performed for 30 min in 50 % formamide/1X SSC, followed by two 0.2X SSC washes for 15 min each. Sections were then incubated with alkaline phosphatase-conjugated sheep anti-DIG (1/1000, Roche) antibody for 2 h. Hybridized probes were visualized by color reaction with nitro-blue tetrazolium (NBT) and 5-bromo 5-chloro-3-indolyl phosphate (BCIP) overnight at 4 °C. Slides were counterstained with Nuclear Fast Red and mounted in a Vectamount mounting medium (Vector Laboratories, Burlingame, USA). In situ analysis was carried out by one “blinded” reader and one non-“blinded” reader, yielding comparable results. No signal was detected using scrambled control probes.
Western Blot
For protein extraction, muscles were homogenized in RIPA lysis buffer containing 150 mM NaCl, 50 mM Tris–HCl pH 7.4, 1 % Nonidet-P40, 1 % glycerol, 1 mM EDTA and protease inhibitors using the Precellys (Ozyme, France) (2 × 10 s, 6500 rpm). Homogenates were centrifuged at 10,000 g to pellet debris and supernatants were centrifuged at 20,000 g (45 min, 4 °C). Protein concentration was determined using a BCA protein assay (Sigma-Aldrich). Fifty μg of proteins of tissue homogenate were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on 4–12 % polyacrylamide gels (NuPage, Life Technologies, Illkirch, France) and electroblotted onto nitrocellulose membranes (Protran BA 83, GE Healthcare) using a Bio-Rad® liquid blotting system at 30 mA for 2 h. The membranes were blocked using 50 % Blocking Buffer (Odyssey®, Li-Cor Biosciences, Lincoln, NE, USA) in PBS (1 h, room temperature) and incubated overnight at 4 °C with primary antibodies against myosin heavy chain (MHC) MF-20 (1/1000, Developmental Studies Hybridoma Bank /DSHB, Iowa City, IA, USA), α-actinin (1/1000, Sigma-Aldrich), MYH7 (1/5000, Abcam, Cambridge, MA, USA). After washing with Tween 0.1 % in PBS, the blots were incubated with horseradish peroxidase-conjugated or fluorophore-conjugated anti-mouse and anti-rabbit secondary antibody. After washing, the samples were coverslipped with Mowiol Mounting Medium (Calbiochem EMD Biosciences, San Diego, CA, USA) and scanned with a blue 488 nm argon ion laser using the C1 inverted Nikon TE-2000 laser scanning confocal microscope (Nikon, Champigny, France). Equal protein loading was checked through α-actinin labelling and Ponceau S staining of the membranes.
Discussion
miRNAs are considered as integral components of the regulatory circuitry for myogenesis, even if their full role in muscle growth and development remains to be elucidated [
13‐
15,
18]. Numerous studies provide increasing evidence for the involvement of miRNAs in myopathies, and particularly in muscular dystrophies [
22,
23,
35]. It has been recently reported that miRNAs are promising biomarkers for monitoring disease progression [
28,
36,
37]. In this regard, several serum miRNAs have been identified as dysregulated in GRMD dogs, using a high-throughput miRNA sequencing screening [
28]. Based on the clinical relevance of the animal model used, these results have allowed authors to select these miRNAs as candidate biomarkers for DMD patients. In addition, it is increasingly acknowledged that miRNAs could represent useful tools in the assessment of experimental therapies to cure muscle diseases [
25,
28,
38]. Nevertheless, further investigations need to be conducted to identify the role of these dysregulated miRNAs in muscle pathophysiology.
Up to now, most of the results presented on muscle miRNAs have been obtained from the
mdx mouse model, which is known to show limitations for the study of pathogenetic mechanisms and therapeutic trials specific to DMD. For this reason, we aim at establishing, for the first time, a description of miRNA dysregulations in GRMD dog skeletal muscle based on a dedicated set: miR-1, miR-133a, miR-206, miR-222 and miR-486. In accordance with previous observations made in the
mdx mouse model and DMD patients [
10,
22,
25,
26], we find that miR-222 and miR-486 exhibit a marked up-regulation and a down-regulation in 9-month-old GRMD dog muscle, respectively. On the contrary, RT-qPCR performed on
Biceps femoris muscle extract fails to reveal any dysregulation of miR-206, in contrast with the previously described up-regulation in both the
mdx mouse model and DMD patients [
10,
22,
25,
26]. Nevertheless, up-regulation of miR-206 is not observed in all dystrophic muscles. Indeed, McCarthy et al. demonstrated that miR-206 is overexpressed in the most severely affected
mdx muscles, i.e. the diaphragm, but not in the hindlimb [
27]. In addition, Yuasa et al. showed a decreased expression of this miRNA in the
CXMDJ tibialis anterior muscle compared to the control [
26]. Our results support Yuasa’s hypothesis that increased expression of miR-206 in
mdx muscle may reflect active and efficient regeneration, whereas its decreased expression in
CXMDJ muscle may illustrate relatively exhausted regeneration potential [
26]. In the present study, we use
in situ detection to obtain original information concerning the muscle tissue distribution of the miRNAs, thus improving the characterization of their tissue expression. Combined
in situ hybridization and MyHCd labelling demonstrate that miR-206 is histologically related to muscle fibre regenerative processes in GRMD dog, being mainly expressed in newly formed fibres. Interestingly, this distribution has been previously reported in
mdx muscles [
10,
26] that have considerable regenerative capacity [
39,
40].
While we describe here the expression patterns of a miRNA subset, further studies are required to understand the implication of miRNAs in the pathophysiology of GRMD dog. In this paper, we show that a continuous cyclosporine-based immunosuppressive regimen maintained over a period of 5 months does not lead to a major modification of the investigated miRNAs levels, except for miR-206 that tends to increase. This result highlights a selective impact of immunosuppression treatment on the expression levels of miRNAs, thus strongly suggesting that the immunosuppressive component must be considered in the assessment of allogeneic cell-based preclinical studies requiring the use of immunosuppression [
41].
In the present study, we attempt to determine whether the systemic delivery of MuStem cells, which increases muscle regenerative activity and stabilizes the clinical status of GRMD dogs [
29], can concomitantly affect the expression levels of miRNAs that are able to modulate key cellular processes at a post-transcriptional level. This hypothesis seems particularly interesting because the observed clinical and tissue benefits following MuStem cell infusion are linked to a low dystrophin protein level as well as a limited percentage of dystrophin-positive fibres, clearly evoking the implication of other molecular pathways [
29,
30]. Firstly, it is surprising that the expression levels of miR-206 and miR-486 (two miRNAs known to be implicated in the regenerative process) are not up-regulated in transplanted GRMD dogs. It could be hypothesized that the increased regenerative potential revealed in GRMD
MuStem dogs 4 to 5 months after transplantation is not sufficient to be associated with a differential expression of miR-206. Secondly, we demonstrate an up-regulation of miR-133a and miR-222 expression after systemic delivery of MuStem cells. Interestingly, changes of these miRNAs are reported to be implicated in the disruption of sarcomere organization [
20,
42]. Expression of miRNA-222 in myoblasts induces myogenin expression followed by inhibition of sarcomeric protein accumulation. Our finding on the down-expression of two sarcomeric proteins MYH7 and MHC in GRMD
MuStem muscle suggests that miR-133a and miR-222 could be involved in the remodelling of the sarcomeric assembly, thus preventing the accumulation of sarcomeric component aggregates observed in dystrophic muscle. Moreover, the pathway analysis performed to provide functional annotation based on KEGG terms (DIANA-miRPath) shows an enrichment of miR-133 in many pathways linked to ubiquitin mediated proteolysis as well as regulation of the actin cytoskeleton. Also, this indicates that miR-222 is involved in the molecular pathways linked to the cell cycle, the insulin signalling pathway and ubiquitin mediated proteolysis.
These observations corroborate our previous study [
30] in which we demonstrated that systemic administration of MuStem cells greatly enhances ubiquitin-mediated protein degradation and induces insulin resistance in skeletal muscle.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
FR designed and performed the miRNA quantification experiments, drafted the manuscript and prepared the figures. CB designed and carried out the in situ hybridization, performed the miRNA quantification experiments, drafted the manuscript and prepared the figures. TL performed clinical follow-up of the dogs, tissue sampling and described the in situ hybridization experiments. LD directed and interpreted the immunofluorescence and laser scanning confocal analysis experiments. ML and HG performed the tissue sampling, imunohistochemistry experiments and morphometric analyses. KR and LG designed the set of experiments, analysed the data and critically revised the manuscript. All authors contributed to the final draft of the manuscript and approved the final version.