All experimental procedures were approved by the Animal Care Committee of the Gwangju Institute of Science and Technology (approval number: GIST-2020-008). C57BL/10ScSn (Dmd^mdx/Utrn^tm1Jrs) mice were purchased from The Jackson Laboratory (Bar Harbor, ME, #016622). We maintained the heterozygous mdx/utrn (±) mouse line by mating two male and female heterozygous mdx/utrn (±) mice and subsequent genotyping of every generation. To ensure an authentic dystrophic cardiomyopathic model, only aged mdx/utrn (±) mice (13–14-month-old; weight, approximately 35 g) were included in our study.

Genomic DNA was obtained from each mouse tail using the DNeasy Blood and Tissue kit (Qiagen, #69504, MD, United States) following the manufacturer’s protocol. Standard PCR was performed using primers provided by Jackson Lab protocol #27099. Primer 9290: 5′-CTG AGT CAA ACA GCT TGG AAG CCT CC-3′ (utrn common reverse); primer 9291: 5′-TTG CAG TGT CTC CCA ATA AGG TAT GAA C-3′ (utrn wild type forward); primer 9292: 5′-TGC CAA GTT CTA ATT CCA TCA GAA GCT G-3′ (utrn mutant forward). The WT and mut utrn PCR product sizes were 646 and 450 bp, respectively. Agarose gel electrophoresis was performed to confirm the individual mouse genotype, and only heterozygote mice, which had two PCR products using the primer sets above, were used in our study.

Recombinant AAV.9 was produced by the transfection of HEK293T cells. AAV.9 particles suspended in culture media were precipitated with ammonium sulfate and purified by ultracentrifugation on an iodixanol gradient. The particles were concentrated using a Vivaspin centrifugal concentrator (Merck, #Z614645). In this step, iodixanol was diluted using lactated Ringer’s solution via multiple washes. The AAV.9 titer was determined by quantitative real-time PCR and SDS-PAGE. AAV.9-Control (Con) and -CCN5 were injected into the tail vein of WT or mdx/utrn (±) mice. Mice received 5 × 10¹¹ viral genomes and were analyzed 4 weeks later. AAV.9-Con encodes the virus-like particle (VLP) gene, which has no effect on fibrosis or cardiac function. AAV.9-CCN5 encodes the human CCN5.

Cardiac tissues from mice were cryopreserved with Tissue-Tek OCT Compound (Sakura, #4583) and sectioned at 6 μm. Tissues were fixed with 4% paraformaldehyde and stained with Masson-Trichrome (Sigma, #HT15-1KT). The fibrotic areas were visualized using a Zeiss Axiophot microscope.

Fibrotic tissue was stained blue and non-fibrotic tissue was stained red. The percentage of fibrotic area was calculated as the ratio of the area of fibrosis to the total area of the section using the ImageJ plug-in software (NIH, United States).

Cardiac fibroblasts (FBs) and myocytes were isolated from 13–14-month-old WT or mdx/utrn (±) mice. Hearts were immediately resected and retrogradely perfused with a Langendorff system filled with Tyrode’s buffer (137 mM NaCl, 5.4 mM KCl, 1 mM MgCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4) for 3 min at 37°C. Next, the buffer was switched to an enzyme solution mixture of 250 U/mL type II collagenase (Worthington, # LS004174) and 60 U/mL hyaluronidase (Worthington, # LS002594) and perfused for 20 min. The digested hearts were transferred to a Petri dish with 5% bovine serum albumin containing Tyrode’s buffer. The cardiac tissue was separated and filtered using a 100 μm pore size strainer (Fisher Scientific, # 352098). The filtered fraction was centrifuged at 50 × g for 1 min. The supernatant, which contained cardiac FBs, was centrifuged at 250 × g for 15 min and resuspended in DMEM (HyClone, # C838R16) containing 10% fetal bovine serum and 1% Penicillin-Streptomycin antibiotics (Thermo Fisher, United States). The cells were seeded in a culture dish; after 4 h, the culture medium was replaced with fresh medium. The precipitated cardiomyocyte fraction was resuspended in phosphate buffered saline and seeded in a 5 μg/mL laminin-coated culture dish. PBS was replaced with M199 medium containing 0.1% bovine serum albumin, 1 × insulin-transferrin-selenium (ITS), 10 mM BDM, 1 × CD lipid (chemically defined lipid concentrate), and 1 × penicillin/streptomycin after 1 h. Cells were incubated at 37°C in a 5% CO₂ incubator.

The isolated cardiomyocytes were plated on laminin-coated coverslips. The culture medium was changed to fresh medium after 1 h and incubated at 37°C in a 5% CO₂ incubator. The next day, cardiomyocytes were treated with 1 μM Fura-2 AM for 10 min before mounting. The cover slip mounted with cardiomyocytes attached was mounted on the stage of an inverted microscope (Nikon Eclipse TE-100F) and perfused (approximately 1 mL/min at 25°C) with Tyrode’s buffer [137 mM NaCl, 5.4 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM glucose, 0.5 mM Taurin, and 10 mM HEPES (pH 7.4)]. Cardiomyocytes were field-stimulated at a frequency of 1 Hz and 40 V using a stimulator (Grass Technologies, United States). The cardiomyocytes were displayed on a monitor screen using an IonOptix MyoCam camera, and changes in cell length during shortening and lengthening were analyzed using IonWizard software.

A dual-excitation spectrofluorometer (IonOptix Inc., United States) was used to record the fluorescence measurements. The cardiomyocytes, mounted on the stage of an inverted microscope, were exposed to light from a 360 or 380 nm wavelength-filtered 75 W halogen lamp. Fluorescence emissions were detected after initial excitation at 360 nm for 0.5 s and then at 380 nm for the duration of the recording protocol. The 340/380 ratio of the fluorescence intensity (FFI) of the Fura indicated the intracellular Ca²⁺ concentration.

Ketamine (100 μg/g) was injected intraperitoneally to anesthetize mice. Short-axis 2D images and M-mode tracings were recorded at the papillary muscle level to measure fractional shortening (FS), left ventricular internal dimension in diastole (LVIDd), and left ventricular internal dimension in systole (LVIDs) (GE Vivid 7 Vision, United States).

A treadmill test was performed as previously described, with some modifications (25). Prior to beginning measurements, WT and mdx/utrn (±) mice were placed on an unmoving treadmill for 2 min for adjustment and then run at 4 m/min for 2 min and 8 m/min for 2 min to warm up. The main exercise was performed at a speed of 12 m/min, until the mice were exhausted. Exhaustion was determined when the mice stopped exercising and did not respond to electrical stimuli for 3–5 s. The aversive electric stimuli were set at 3 Hz, 1.22 mA to maintain the mice running. There was no inclination during this test. The total exercise distance was calculated by multiplying the exercise time and velocity. At the end of the test, mice were allowed to rest for 24 h and the harvested for molecular analysis.

The forearm grip strength of the WT and mdx/utrn (±) mice was assessed using an automated grip strength meter (Bioseb, BIO-GS3). The mice grasp a grid with their forepaws instinctively and try to hold while being pulled backward by the tail, releasing when they are unable to maintain grip. Five measurements within 2 min were taken from each animal at 1 week interval after the AAV.9 virus injection.

Cardiac tissues, thigh skeletal muscle, and cell lysates were solubilized in RIPA buffer (50 mM Tris, 150 mM NaCl, 0.1% SDS, 1% Triton X-100, pH 8.0) mixed with protease inhibitor cocktail set III (Merck Millipore, #535140). Lysate concentrations were quantified using a Pierce BCA Protein Assay Kit (Thermo Scientific, #23227). Lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes (Merck Millipore, #IPVH00010). Transferred blots were blocked with 5% non-fat skim milk and incubated with antibodies against mouse monoclonal CCN5 (1:1,000, Sigma, #WH0008839M9), mouse monoclonal α-SMA (1:1,000, Sigma-Aldrich, #A5228), goat polyclonal collagen I (1:1,000, Abcam, ab34710), rabbit polyclonal SERCA2a (1:1,000, a custom antibody from 21st Century Biochemicals), rabbit polyclonal RyR2 (1:1,000, Alomone Lab, ARR-002), rabbit monoclonal NCX1 (1:1,000, Abcam, EPR12739), rabbit polyclonal p-PLN (1:1,000, Badrilla, A010-12AP), mouse monoclonal t-PLN (1:1,000, Badrilla, A010-14), and mouse monoclonal α-tubulin (1:3,000, Santa Cruz, #sc-8035) for 12–16 h at 4°C. After washing the blots with Tris-buffered saline containing 0.1% Tween 20 (TBS-T), they were incubated with secondary antibodies conjugated with horseradish peroxidase (Thermo Fisher Scientific, #31430) and washed again. The band signal was developed using chemiluminescence solution (Millipore, #WBKLS0500). ImageJ plug-in software was used to quantify the western blot results.

The mRNA expression level of utrophin was determined by real-time PCR using the QuantiTect SYBR Green real-time PCR Kit (Qiagen, #204243). Total RNA was isolated from cardiac tissues using TRIzol reagent (Invitrogen, #15596), according to the manufacturer’s instructions. Reverse transcription was performed at 50°C for 20 min, and cDNA was amplified in 20 μL reaction volumes using 10 pmol of primers for 37 cycles: 94°C for 10 s, 57°C for 15 s, and 72°C for 5 s. To calculate the relative abundance of the mRNAs, 18S rRNA was used as an internal control. The primer sequences used in this study were as follows: utrophin, 5′-TCA TGC TAG CCT GGA CCA TTT T-3′ (forward) and 5′-CAC TGA TGG GTG GTT TCC CA-3′ (reverse); 18S rRNA, 5′-TAA CGA ACG AGA CTC TGG CAT-3′ (forward) and 5′-CGG ACA TCT AAG GGC ATC ACA G-3 (reverse).

The utrophin (UTRN) promoter was cloned into a pEZX-PG02.1 vector from a commercial source (GeneCopeia). UTRN promoter DNA (1 μg) and pcDNA3.1-CCN5-myc-his DNA (1 μg) were mixed with 3 μg polyethylenimine (PEI) and incubated for 20 min at 25°C. HEK293T cells were treated with DNA-PEI complexes, and after 4 h the culture medium was replaced with fresh medium. After 72 h, the culture media and lysates were collected and used to detect luciferase activity using a Pierce Gaussia Luciferase assay kit (Thermo Fisher, #16160). The luciferase signal was analyzed using a GloMax microplate reader. GABPα DNA was used as a positive control for the UTRN promoter assay.

Student’s t-test and one-way analysis of variance (ANOVA) were used for statistical analyses to determine the significance of the data. Asterisks (*p < 0.05) or double asterisks (^**p < 0.01) indicate significance. Data in the figures represent the mean ± standard deviation.

Two AAV.9 viral vectors with a CCN5 cassette (AAV.9-CCN5) and control virus (AAV.9-Con) were generated to examine the effects of CCN5 on cardiac fibrosis. We used at least 13-month-old mdx/utrn (±) mice to ensure the establishment of cardiac fibrosis and clinical relevance prior to CCN5 gene transfer. Mice received AAV.9-CCN5 (1E11 vg/mouse) through the tail vein and were analyzed 8 weeks later (Figure 1A). First, cardiac tissues were subjected to histological analysis. Heart sections were stained with Masson’s trichrome reagent to analyze CF. Fibrotic areas were significantly developed in the mid- and sub-epicardium within the inferior and lateral walls, in line with the literature (26). AAV.9-Con did not affect fibrotic development (Figures 1B,C). However, a significant decrease in fibrosis within the walls was observed in AAV9-CCN5-administered mice (Figures 1B,C). Molecular signatures of cardiac fibrosis were also evaluated by western blot analysis, which supported this observation. For example, α-SMA, a TGF-β-induced marker of myofibroblast differentiation, was exceptionally inhibited by CCN5 gene transfer (Figures 1D,E). Furthermore, levels of major calcium regulators, such as SERCA2a, RyR2, NCX1, and phospholamban (PLN, phosphor, and total forms) in the heart were measured in the same samples. SERCA2a expression, a hallmark of calcium regulation in the heart, was markedly preserved in CCN5 gene-transfected mice (Figures 1D,E). In addition, phosphorylated PLN levels were significantly restored in the CCN5 gene transfer group. Finally, RyR2 expression was not altered in this model, although NCX1 expression was downregulated in CCN5 injected group. Moreover, the antifibrotic effect of CCN5 was confirmed in isolated cardiac fibroblasts in vitro by western blot analysis, and representative profibrotic markers, such as α-SMA, collagen I, and TGF-β, were significantly reduced by CCN5 gene transfer (Supplementary Figures 1A,B). These data prove that CCN5 ameliorates cardiac fibrosis in mdx/utrn (±) mice.

Restoration of SERCA2a levels results in enhanced calcium handling, which leads to increased cardiomyocyte contractility (27, 28). Therefore, we examined the effect of CCN5 overexpression on cardiomyocyte contractility in vitro. We also assessed cardiac function using echocardiography in vivo, since cardiac fibrosis negatively affects cardiac function (29–32).

Both cardiac fibroblasts and cardiomyocytes were isolated for further measurements. For this experiment, a modified Langendorff technique was applied to isolate both cell types from the hearts of mdx/utrn (±) mice, and Ca²⁺ transients and cardiomyocyte contractility were assessed using the IonOptix system, as described in section “Cardiomyocyte Contractility Analysis and Intracellular Ca²⁺ Transient Measurement.”

CCN5 gene transfer resulted in significantly enhanced Ca²⁺ transients and contractility in isolated cardiomyocytes compared to those in control viral vector-delivered mice (Figures 2A–D). Furthermore, the cardiac FB fraction was analyzed by western blotting, and several profibrotic markers were significantly reduced by CCN5 overexpression, as described previously (Supplementary Figure 1).

To investigate in vivo cardiac function, echocardiography was performed 8 weeks after gene transfer (Figure 1A). Systolic function indices, including fractional shortening (FS) and ejection fraction (EF), were significantly enhanced, and LVIDd and LVIDs were normalized by CCN5 gene transfer (Figures 2E,F and Supplementary Table 1). These data suggest that CCN5 reduces cardiac dysfunction in mdx/utrn (±) mice by restoring calcium regulation and cardiomyocyte contractility both in vitro and in vivo.

One of the primary symptoms in patients with DMD is muscle weakness throughout the body, which begins in the lower body around the thighs and pelvis (33–35). In animal models of DMD, muscle weakness has also been observed in multiple reports (36–38). In our aged mouse model, this phenomenon is heightened, and animals are unable to intake food and water properly due to their inability to cling onto the food dispenser. Therefore, food and water were placed at the bottom of the cage to make it easily accessible. However, aged mice with CCN5 gene transfer showed better mobility. Therefore, we performed exercise performance tests using a treadmill and a grip strength instrument for accurate measurements (Figure 3A) (25, 39). mdx/utrn (±) mice were exhausted at half the distance compared to the WT control, whereas mice with CCN5 gene transfer exhibited remarkably enhanced running ability (Figure 3B) as well as grip strength (Figure 3C). In addition, molecular signatures analyzed by western blotting and qRT-PCR after CCN5 gene transfer supported these observations. As shown in Figures 3D,E, the transcriptional and translational levels of α-SMA, a myofibroblast marker, were significantly reduced. Surprisingly, utrophin expression was also substantially elevated in skeletal muscle with CCN5 gene transfer, which implies that CCN5 modulates not only tissue fibrosis but also the transcription of utrophin. This result led us to further characterize the relationship between CCN5 and utrophin expression.

Our results in section “CCN5 Enhances Exercise Performance of mdx/utrn (±) Mice” and previous literature imply that CCN5 can act as a transcriptional regulator in the nucleus (Supplementary Figure 3) (40). Our results suggest that CCN5 may modulate gene expression of certain structural molecules in cells. We identified utrophin as a novel target of CCN5 using gene expression profiling (data not shown).

To test this hypothesis, we observed utrophin levels in mdx/utrn (±) mice 8 weeks after CCN5 gene transfer. We confirmed mdx/utrn (±) genotypes following the Jackson Lab protocol, as described in section “Materials and Methods” (Supplementary Figure 2). We analyzed utrophin expression by western blotting and qRT-PCR.

Utrophin expression was significantly induced at both protein and mRNA levels (Figures 4A–C). In contrast, dystrophin expression was unaltered in all animals (Figures 4A,B). Furthermore, to confirm whether this effect was directly mediated between CCN5 and the target gene promoter, we performed a luciferase assay using the utrophin promoter with the positive control, GABPα, which is a previously reported transcriptional regulator of utrophin (Figure 4D). Our data suggest that CCN5 directly upregulates utrophin expression at the transcriptional level.