Low-dose MyoAAV 2A-mediated delivery of engineered micro-utrophin achieves pan- muscle tissue distribution with elevated muscle function in Duchenne muscular dystrophy Authors: Na Li, Zhuo Li, Zilian Zhang, Zhikai Xiahou, Hong Xiao, Shiwen Wu, Fuqiang Xu, Qing Liu, Yang Wu, Yafeng Song Published: 2026 ABSTRACT BackgroundThe pathogenesis of Duchenne muscular dystrophy (DMD) is driven by a deficiency of functional dystrophin. Although gene therapy is a promising strategy, current approaches face substantial challenges in balancing therapeutic efficacy with toxicity, largely owing to the requirement for high-dose adeno-associated virus (AAV) administration.MethodsWe employed a low-dose strategy in neonatal mdx mice, using a muscle-targeted MyoAAV 2 A capsid to deliver a rationally designed, codon-optimized micro-utrophin construct (ΔR4–R22/ΔC + H3/H4) at 1 × 10¹² vg/kg via intraperitoneal injection. Four weeks post-injection, micro-utrophin expression and localization, as well as γ- and α-sarcoglycan recruitment, were assessed by western blotting and immunofluorescence. Vector genome distribution was quantified by real-time quantitative PCR. Tissue histology was evaluated using Evans blue dye uptake, H&E staining, and Masson’s trichrome staining. Muscle function was measured by grip strength and treadmill exercise tests. Serum biomarkers of cardiac injury, heart failure, and systemic inflammation were detected by ELISA.ResultsRobust micro-utrophin expression was detected in skeletal muscles, diaphragm, and myocardium, with minimal off-target expression in the liver. Sarcolemmal localization and functional recruitment of γ- and α-sarcoglycan confirmed restoration of the dystrophin–glycoprotein complex. In treated animals, serum creatine kinase levels, Evans blue dye uptake, and central nucleation in multiple muscles were significantly reduced, accompanied by improvements in relative grip strength, running time, and distance to exhaustion. No cardiac or hepatic toxicity was observed, as indicated by unchanged cTnI, NT-proBNP, ALT, and AST levels. Additionally, serum IL-6 and C5a concentrations were significantly decreased, and fibrosis and inflammation in the diaphragm and myocardium were markedly attenuated.ConclusionsThese findings demonstrate that combining a functional micro-utrophin construct with an advanced muscle-tropic MyoAAV 2 A capsid constitutes a promising strategy to overcome the efficacy–toxicity trade-off that has hindered the clinical translation of DMD gene therapies.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12967-026-08483-6. Introduction Duchenne muscular dystrophy (DMD) is a devastating X-linked neuromuscular disorder caused by a deficiency of functional dystrophin and affects approximately 1 in 3,500 to 5,000 male births. The absence of dystrophin results in a considerable reduction of the dystrophin–glycoprotein complex (DGC), a dynamic transmembrane multi-protein complex essential for maintaining membrane integrity during muscle contraction [1, 2]. The loss results in membrane destabilization, and subsequent myofiber degeneration. Consequently, affected individuals typically lose ambulatory ability between the ages of 8 and 12, and often succumb to respiratory or cardiac failure by their second or third decade of life [3]. Recent years have witnessed encouraging results from clinical trials of gene transfer therapies designed to mitigate DMD progression through the delivery of a functional, truncated form of dystrophin to affected muscles [4]. These outcomes contributed to the accelerated approval of the first gene therapy, delandistrogene moxeparvovec (ELEVIDYS®). The therapy has received traditional approval for ambulatory patients aged 4 years and older with a confirmed DMD mutation, as well as accelerated approval for non-ambulatory individuals meeting the same age and genetic confirmation criteria [5]. However, fatal cases of acute liver failure have been reported following ELEVIDYS administration. Therefore, the overall benefit-risk profile of ELEVIDYS requires further validation through confirmatory trials. Current leading DMD gene therapies—including ELEVIDYS and two other candidates in clinical trials—rely on high-dose systemic administration (1–3 × 1014 vg/kg) of AAV9 or AAVrh74 vectors carrying engineered micro-dystrophins with optimized spectrin-like repeats and hinge domains [4, 6, 7]. Although these constructs restore functional dystrophin expression, dose-dependent toxicities have been observed, such as systemic inflammatory responses (grade ≥ 3), acute kidney injury, and cardiopulmonary complications, particularly at doses exceeding 2 × 1014 vg/kg [8, 9]. Studies in large animal models further suggest that intravenous AAV doses above 7.5 × 1013 vg/kg can trigger toxic reactions [10–12]. Another major concern is immunogenicity, as many DMD patients carry multi-exon deletions (e.g., Δ45–50), which render the delivered micro-dystrophin susceptible to recognition as a foreign antigen, potentially eliciting autoimmune myositis. Therefore, there is an urgent need to develop safer strategies that minimize both AAV-related toxicity and immune responses. MyoAAVs, which are a class of engineered AAV capsids developed through directed evolution, demonstrate significantly enhanced muscle-specific transduction [13]. In a comparative preclinical study evaluating MyoAAV 2A and AAV9 for delivery of mini-dystrophin in mdx mice, MyoAAV 2A demonstrated superior performance, with 12- to 46-fold higher vector copy numbers and 7.6- to 15-fold elevation in target mRNA levels. These molecular advantages translated into functional benefits, including significantly improved specific force recovery and enhanced protection against muscle damage. A notable safety advantage was observed in biodistribution profiles, with mice treated with MyoAAV 2A showing 2.5-fold lower vector genomes in liver tissue compared to the AAV9 group [14, 15]. Collectively, the high myotropic specificity, minimized off-target transduction, and favorable safety profile establish MyoAAV 2A as a promising vector for delivering DMD gene therapy in mdx mice, with potential translational advantages in both efficacy and safety. Utrophin, an autosomal paralogue of dystrophin sharing approximately 80% structural and functional similarity, is constitutively expressed in diverse tissues of both healthy individuals and DMD patients [16, 17]. Multiple studies have demonstrated that the first-generation micro-utrophin (ΔR4–R21/ΔC) can recruit the DGC, improve cardiac and skeletal muscle function, and maintain a non-immunogenic profile [18–24]. Based on these premises, we sought to evaluate the therapeutic potential of a newly engineered micro-utrophin delivered via MyoAAV 2A at a low dose. We hypothesized that systemic administration of this optimized construct at 1 × 1012 vg/kg would ameliorate muscle pathology and improve functional outcomes in mdx mice. The findings from this study are expected to provide critical preclinical evidence regarding the suitability of this strategy for further evaluation in large animal models of DMD. Results Rationally designed and codon-optimized micro-utrophin Structural analyses of spectrin-like repeats (SRs) in dystrophin and utrophin, combined with phenotypic studies of in-frame deletions and duplications in Becker muscular dystrophy (BMD) patients, reveal that amino acid side-chain interactions between adjacent repeats form an interlocking interface crucial for rod domain stability. Disruption of native SR sequences compromises local structural integrity, leading to focal weakening of the rod domain [25, 26]. To minimize such structural vulnerabilities, we designed a novel micro-utrophin construct that retains the N-terminal actin-binding domain (NT), the first three SRs, and the cysteine-rich (CR) domain, while introducing deletions between hinge 2 (H2) and hinge 4 (H4). In this design, hinge 1 (H1) serves as a linker between the NT and the first three SRs. Critically, to enhance mechanical resilience and improve domain continuity, we replaced the previously used H2 linker with a novel hybrid hinge. This hinge was created by fusing the N-terminal segment of hinge 3 (H3) to the C‑terminal segment of hinge 4 (H4), thereby bridging the first three SRs to the CR domain. The resulting construct was designated ΔR4–R22/ΔC + H3/H4 (Fig. 1a). Notably, H3 adopts a flexible, loop-rich conformation that is less prone to mechanical failure compared to the more structured H2 employed in our prior variant (ΔR4–R21/ΔC) [25]. The inherent structural features of H3 and H4 further predispose the chimeric hinge to adopt an N‑terminal α‑helix followed by a C‑terminal β‑sheet (Supplementary Fig. 1). Through systematic screening, we identified an optimal hinge configuration comprising the first 30 amino acids of H3 and the last 7 amino acids of H4 (Fig. 1b). Finally, the micro-utrophin transgene was synthesized based on the wild‑type utrophin and codon‑optimized based on highly expressed myosin isoforms in muscle to maximize translational efficiency [27]. Fig. 1Schematic view of the structural domains in utrophin and micro-utrophin and the experimental timeline. a. The domains contained in full-length utrophin (utrophin-FL) and micro-utrophin. Utrophin-FL contains an N-terminal actin-binding domain (NT), a central rod-like region composed of 22 spectrin-like repeats (SRs) dispersed by four hinge regions (H1-H4), a cysteine-rich (CR) domain and a carboxy-terminal domain (CT). Micro-utrophin retains the NT domain, H1 and the first three SRs in the central rod, the total CR domain, but deletes the CT domain. The truncated central rod and CR domain are held together by a novel hinge region formed by the fusion of H3 and H4. b. Structure of both the utrophin-FL and micro-utrophin predicted by AlphaFold 3. The dark purple segment corresponds to the NT, the light-blue region comprises SR1, SR2, and SR3, while subsequent SRs are depicted in light green. Notably, the crimson-colored region contains the WW domain and ZZ domain within the CR region, whereas utrophin-FL incorporates both WW and ZZ domains in addition to contributions from the CT. In addition, a novel helical structure formed by the engineered hinge region is highlighted in green. c. The packaging process of MyoAAV 2 A–micro-utrophin and experimental timeline. MyoAAV 2 A–micro-utrophin was packaged in 293T cells using three plasmids, namely pAAV-micro-utrophin, pAAV-Rep/Cap and pAd-Helper Schematic view of the structural domains in utrophin and micro-utrophin and the experimental timeline. a. The domains contained in full-length utrophin (utrophin-FL) and micro-utrophin. Utrophin-FL contains an N-terminal actin-binding domain (NT), a central rod-like region composed of 22 spectrin-like repeats (SRs) dispersed by four hinge regions (H1-H4), a cysteine-rich (CR) domain and a carboxy-terminal domain (CT). Micro-utrophin retains the NT domain, H1 and the first three SRs in the central rod, the total CR domain, but deletes the CT domain. The truncated central rod and CR domain are held together by a novel hinge region formed by the fusion of H3 and H4. b. Structure of both the utrophin-FL and micro-utrophin predicted by AlphaFold 3. The dark purple segment corresponds to the NT, the light-blue region comprises SR1, SR2, and SR3, while subsequent SRs are depicted in light green. Notably, the crimson-colored region contains the WW domain and ZZ domain within the CR region, whereas utrophin-FL incorporates both WW and ZZ domains in addition to contributions from the CT. In addition, a novel helical structure formed by the engineered hinge region is highlighted in green. c. The packaging process of MyoAAV 2 A–micro-utrophin and experimental timeline. MyoAAV 2 A–micro-utrophin was packaged in 293T cells using three plasmids, namely pAAV-micro-utrophin, pAAV-Rep/Cap and pAd-Helper Rationally designed micro-utrophin delivered via MyoAAV 2 A recruited the functional expression of DGC in dystrophic muscle of mdx mice To explore whether therapy-limiting toxicity in DMD could be circumvented, we employed a muscle-tropic vector, MyoAAV 2 A, to deliver a codon-optimized micro-utrophin transgene. A dose-ranging study was first conducted in neonatal mdx mice, with intraperitoneal injections at doses spanning from 1.0 × 10¹¹ to 1.0 × 1014 vg/kg. Ten days after delivery, micro-utrophin expression in skeletal muscle, diaphragm, myocardium, and liver was assessed by western blotting (Supplementary Fig. 2). Notably, the expression level observed at 1.0 × 10¹² vg/kg appeared to be roughly comparable to those seen at the higher doses of 1.0 × 10¹³ and 1.0 × 10¹⁴ vg/kg, while possibly offering a reduced risk of toxicity. Therefore, to balance efficacy and safety, this dose was selected for further evaluation. To more rigorously validate its therapeutic potential, we systematically assessed the effects of MyoAAV 2 A‑delivered micro-utrophin at 1.0 × 10¹² vg/kg four weeks post-delivery (Fig. 1c). At this time point, micro-utrophin expression, along with DGC components α‑sarcoglycan (α-sg) and γ-sarcoglycan (γ-sg), was evaluated in dystrophic muscle. In randomized, blinded studies, western blotting analysis confirmed that micro-utrophin expression was detected exclusively in MyoAAV 2 A–micro-utrophin‑treated mice and exhibited a striated muscle-specific biodistribution (Fig. 2). Quantification of western blots showed that the normalized micro-utrophin expression level (target/vinculin ratio) in the diaphragm was 1.04 ± 0.08 arbitrary units, followed by quadriceps femoris (0.88 ± 0.09), tibialis anterior (0.86 ± 0.02), myocardium (0.86 ± 0.04), triceps (0.77 ± 0.06), and gastrocnemius (0.65 ± 0.07) (Figs. 2a–f). In contrast, hepatic expression was lower (0.41 ± 0.03) (Supplementary Fig. 3). Robust micro-utrophin expression was associated with increased γ‑sg levels, particularly in the diaphragm (U = 0, exact p = 0.0571, rg = 1.000), quadriceps femoris (U = 0, exact p = 0.0571, rg = 1.000), and gastrocnemius (U = 0, exact p = 0.0571, rg = 1.000). Although the exact p‑values (0.0571) did not reach the significance level (α = 0.05), the complete separation (U = 0), large effect sizes (rg = 1.000), and Hodges-Lehmann median differences with 95% confidence intervals (provided in Fig. 2 legend) lying entirely above zero across three independent muscles collectively indicate a consistent and biologically meaningful elevation of γ‑sg levels following micro-utrophin expression (Figs. 2a–b, f). Although α-sg expression remained largely unchanged overall, a meaningful increase was observed in triceps brachii (U = 0, exact p = 0.0571, rg = 1.0000) (Fig. 2e). Of note, neither γ-sg nor α-sg was detected in the liver across all groups (Supplementary Fig. 3). Fig. 2MyoAAV 2 A–micro-utrophin mediated muscle-targeted expression of micro-utrophin and thereby recruited the DGC. a Left: western blots of diaphragm (WT + PBS, mdx + PBS, mdx + MyoAAV 2A–micro-utrophin; 4 weeks). Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 2, p = 0.4000 rg = 0.5556, 95% CI for rg = [-0.5556, 1.0000], median difference = 0.1269, 95% CI for median difference = [-0.1336, 0.4136]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 4, exact p = 0.6286, rg = 0.3333, 95% CI for rg = [-0.5000, 1.0000], median difference = -0.0956, 95% CI for median difference = [-0.4262, 0.2767]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 1, exact p = 0.2000, rg = 0.7778, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2857, 95% CI for median difference = [–0.0521, 0.4943]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.5165, 95% CI for median difference = [0.1489, 0.5905]). b Left: western blots of quadriceps femoris from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4761, 95% CI for median difference = [0.3687, 0.7033]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 1, exact p = 0.1143, rg = 0.8333, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1291, 95% CI for median difference = [–0.0132, 0.3202]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2623, 95% CI for median difference = [0.0974, 0.5713], mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.2100, 95% CI for median difference = [0.0266, 0.5188]). c Left: western blots of tibialis anterior from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4840, 95% CI for median difference = [0.2895, 0.7506]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 2, exact p = 0.2286, rg = 0.6667, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1504, 95% CI for median difference = [–0.0107, 0.4360]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 1, exact p = 0.2000, rg = 0.7778, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.3473, 95% CI for median difference = [–0.0305, 0.7252]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 1, exact p = 0.1143, rg = 0.8333, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.6206, 95% CI for median difference = [–0.0777, 1.0797]). d Left: western blots of myocardium from the same groups. Right: quantification of α‑sg ( WT + PBS vs. mdx + PBS:U = 2, exact p = 0.4000, rg = 0.5556, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0914, 95% CI for median difference = [–0.1228, 0.4052]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.4167, 1.0000], median difference = − 0.1478, 95% CI for median difference = [–0.4753, 0.3149]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4473, 95% CI for median difference = [0.1835, 0.6468]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: (U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1090, 95% CI for median difference = [–0.1329, 0.4808]). e Left: western blots of triceps brachii from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4592, 95% CI for median difference = [0.2987, 0.6114]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.5110, 95% CI for median difference = [0.2173, 0.7204]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4679, 95% CI for median difference = [0.3828, 0.5272]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.5000, 1.0000], median difference = 0.3084, 95% CI for median difference = [–0.1808, 0.5429]). f Left: western blots of gastrocnemius from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.3721, 95% CI for median difference = [0.2146, 0.6151]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 2, exact p = 0.2286, rg = 0.6667, 95% CI for rg = [–0.4167, 1.0000], median difference = − 0.1212, 95% CI for median difference = [–0.2152, 0.0075]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2921, 95% CI for median difference = [0.0774, 0.5471]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.3085, 95% CI for median difference = [0.1616, 0.6145]). n = 3 for PBS treated WT and mdx mice. n = 4 for MyoAAV 2 A–micro-utrophin treated mdx mice Immunofluorescence analysis further confirmed sarcolemmal localization of micro-utrophin and enhanced expression of γ-sg and α-sg at the sarcolemma, indicating functional DGC recruitment and structural integration (Figs. 3a–c). To distinguish between exogenous micro-utrophin and endogenous full-length utrophin, we used an antibody recognizing the N‑terminal domain (detects both exogenous and endogenous utrophin) and an antibody targeting the C‑terminal domain (detects only endogenous utrophin, as our micro-utrophin construct lacks the C‑terminus). Using the N‑terminal antibody, strong sarcolemmal signals were observed in treated muscles. Using the C‑terminal antibody, endogenous full-length utrophin remained readily detectable at 4 weeks post‑treatment, with no apparent reduction compared to untreated controls. This may be due to the early time point (4 weeks) analyzed. While a discrepancy was observed between western blotting and immunofluorescence analyses regarding the expression of γ-sg and α-sg in mdx mice, it did not affect the assessment of DGC recruitment. This is because the immunofluorescence signal reflects functional protein localized at the sarcolemma, which is the key determinant for evaluating recruitment. Furthermore, while some interstitial background signal was detected for utrophin by immunofluorescence, it did not interfere with the clear demonstration of micro-utrophin localization at the sarcolemma. This conclusion is supported by negative control experiments, which showed diffuse interstitial background but an absence of the specific sarcolemmal signal ( Supplementary Fig.4). Moreover, future long-term studies will be essential to monitor dynamic changes in endogenous utrophin over an extended treatment period and to elucidate its regulatory mechanisms using more specific detection methods. Fig. 3Muscle-targeted micro-utrophin and elevated γ-sg and α-sg co-localized at the sarcolemma. a Immunofluorescence staining for utrophin using an N-terminal antibody (Utro-N, recognizing both native full-length utrophin and recombinant micro-utrophin, red) and a C-terminal antibody (Utro-C, recognizing only native utrophin, red), with laminin (green) as a basement membrane marker, in diaphragm, quadriceps femoris, tibialis anterior, myocardium, triceps brachii and gastrocnemius from WT mice treated with PBS and mdx mice treated with MyoAAV 2A –micro-utrophin or PBS alone. b Immunofluorescence staining for α-sg (red) and laminin (green) in the above-mentioned six tissues. c Immunofluorescence staining for γ-sg (red) and laminin (green) in the above-mentioned six tissues. Scale bars: 20 μm. WT, wild type MyoAAV 2 A–micro-utrophin mediated muscle-targeted expression of micro-utrophin and thereby recruited the DGC. a Left: western blots of diaphragm (WT + PBS, mdx + PBS, mdx + MyoAAV 2A–micro-utrophin; 4 weeks). Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 2, p = 0.4000 rg = 0.5556, 95% CI for rg = [-0.5556, 1.0000], median difference = 0.1269, 95% CI for median difference = [-0.1336, 0.4136]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 4, exact p = 0.6286, rg = 0.3333, 95% CI for rg = [-0.5000, 1.0000], median difference = -0.0956, 95% CI for median difference = [-0.4262, 0.2767]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 1, exact p = 0.2000, rg = 0.7778, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2857, 95% CI for median difference = [–0.0521, 0.4943]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.5165, 95% CI for median difference = [0.1489, 0.5905]). b Left: western blots of quadriceps femoris from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4761, 95% CI for median difference = [0.3687, 0.7033]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 1, exact p = 0.1143, rg = 0.8333, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1291, 95% CI for median difference = [–0.0132, 0.3202]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2623, 95% CI for median difference = [0.0974, 0.5713], mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.2100, 95% CI for median difference = [0.0266, 0.5188]). c Left: western blots of tibialis anterior from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4840, 95% CI for median difference = [0.2895, 0.7506]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 2, exact p = 0.2286, rg = 0.6667, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1504, 95% CI for median difference = [–0.0107, 0.4360]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 1, exact p = 0.2000, rg = 0.7778, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.3473, 95% CI for median difference = [–0.0305, 0.7252]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 1, exact p = 0.1143, rg = 0.8333, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.6206, 95% CI for median difference = [–0.0777, 1.0797]). d Left: western blots of myocardium from the same groups. Right: quantification of α‑sg ( WT + PBS vs. mdx + PBS:U = 2, exact p = 0.4000, rg = 0.5556, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0914, 95% CI for median difference = [–0.1228, 0.4052]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.4167, 1.0000], median difference = − 0.1478, 95% CI for median difference = [–0.4753, 0.3149]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4473, 95% CI for median difference = [0.1835, 0.6468]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: (U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.1090, 95% CI for median difference = [–0.1329, 0.4808]). e Left: western blots of triceps brachii from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4592, 95% CI for median difference = [0.2987, 0.6114]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.5110, 95% CI for median difference = [0.2173, 0.7204]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4679, 95% CI for median difference = [0.3828, 0.5272]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.4000, rg = 0.5000, 95% CI for rg = [–0.5000, 1.0000], median difference = 0.3084, 95% CI for median difference = [–0.1808, 0.5429]). f Left: western blots of gastrocnemius from the same groups. Right: quantification of α‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.3721, 95% CI for median difference = [0.2146, 0.6151]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 2, exact p = 0.2286, rg = 0.6667, 95% CI for rg = [–0.4167, 1.0000], median difference = − 0.1212, 95% CI for median difference = [–0.2152, 0.0075]) and γ‑sg (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.1000, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.2921, 95% CI for median difference = [0.0774, 0.5471]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0571, rg = 1.0000, 95% CI for rg = [–0.4167, 1.0000], median difference = 0.3085, 95% CI for median difference = [0.1616, 0.6145]). n = 3 for PBS treated WT and mdx mice. n = 4 for MyoAAV 2 A–micro-utrophin treated mdx mice Muscle-targeted micro-utrophin and elevated γ-sg and α-sg co-localized at the sarcolemma. a Immunofluorescence staining for utrophin using an N-terminal antibody (Utro-N, recognizing both native full-length utrophin and recombinant micro-utrophin, red) and a C-terminal antibody (Utro-C, recognizing only native utrophin, red), with laminin (green) as a basement membrane marker, in diaphragm, quadriceps femoris, tibialis anterior, myocardium, triceps brachii and gastrocnemius from WT mice treated with PBS and mdx mice treated with MyoAAV 2A –micro-utrophin or PBS alone. b Immunofluorescence staining for α-sg (red) and laminin (green) in the above-mentioned six tissues. c Immunofluorescence staining for γ-sg (red) and laminin (green) in the above-mentioned six tissues. Scale bars: 20 μm. WT, wild type Rationally designed micro-utrophin delivered via MyoAAV 2A improved muscle function in mdx mice Having assessed the widespread, sarcolemma-localized expression of micro-utrophin across multiple muscle groups, we next sought to determine the vector genome abundance and histological features in MyoAAV 2A–micro-utrophin treated mice compared to PBS injected mdx mice. MyoAAV 2 A–micro-utrophin delivery resulted in markedly higher levels of vector genomes per diploid genome (vg/dg) in skeletal muscles (quadriceps femoris: 11.90 + 8.16; gastrocnemius: 5.35 + 4.42), diaphragm (12.83 ± 8.18), and myocardium (5.63 ± 2.07) than in the liver (0.22 ± 0.12) and all other non‑muscle tissues tested (e.g., spleen, kidney, gonads, lung), where vg/dg levels were uniformly low (Fig. 4a). Consistent with these biodistribution findings, serum CK levels were dramatically increased in mdx + PBS compared to WT + PBS (t = 12.66, p < 0.0001), but were significantly lowered by MyoAAV 2 A–micro-utrophin treatment (t = 2.860, p = 0.0227; Fig. 4b). We also found that MyoAAV 2 A–micro-utrophin treatment significantly reduced EBD infiltration into the gastrocnemius compared to PBS injected mdx controls (U = 0, exact p = 0.0022, rg = 1.000; Fig. 4c), providing direct evidence of improved sarcolemmal integrity and membrane stability. Histological assessment via H&E staining indicated that systemic administration of MyoAAV 2 A–micro‑utrophin also reduced central nucleation in the quadriceps femoris, tibialis anterior, and triceps brachii (each: U = 0, exact p = 0.0286, rg = 1.000; Fig. 4d) (median differences and 95% CIs are provided in the Fig. 4 legend). Notably, the absence of significant effects in the gastrocnemius is likely due to lower micro‑utrophin expression and α‑sg recruitment. To assess functional recovery, grip strength and treadmill endurance tests were performed in a fully blinded and randomized manner. There were no significant differences in body weight between MyoAAV 2 A–micro-utrophin treated and PBS treated mdx mice (t = 0.4608, p = 0.6624; Fig. 4e). Relative grip strength, normalized to body weight, was significantly higher in MyoAAV 2A–micro-utrophin treated mdx mice than in PBS treated mdx mice (t = 4.55, p = 0.0012; Fig. 4f). Running time to exhaustion (mdx + PBS vs. mdx +MyoAAV 2 A–micro-utrophin: t = 4.27, p = 0.0017; Fig. 4g) and total running distance (t = 4.25, p = 0.0018; Fig.4h) were also significantly improved in MyoAAV 2 A–micro-utrophin treated mice. Fig. 4MyoAAV 2 A–micro-utrophin resulted in abundant vector genomes and improved muscle morphology and function. a Quantification of vector genome per diploid genome in various muscles, liver and other tissue of mdx mice injected with MyoAAV 2 A–micro-utrophin, n = 6. b Quantification of serum CK levels. WT + PBS (185.80 ± 10.20) vs. mdx + PBS (4517.00 ± 342.00): Welch’s t = 12.66, p < 0.0001, Cohen’s d = 7.31 (95% CI [4.10, 10.97]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (3417 ± 175.90): Welch’s t = 2.860, p = 0.0227, Cohen’s d = 1.65 (95% CI [0.34, 2.96]). n = 6. c Representative image (left) and quantification (right) of Evans Blue Dye (EBD) and Laminin staining of gastrocnemius from WT mice treated with PBS and mdx mice treated with MyoAAV 2A–micro-utrophin or the same volume of PBS (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0022, rg = 1.000, 95% CI for rg = [0.8750, 1.0000], median difference = 10.6698, 95% CI for median difference = [7.2296, 17.9100]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0022, rg = 1.000, 95% CI for rg = [0.8750, 1.0000], median difference = 5.2267, 95% CI for median difference = [0.3954, 15.4266]). Scale bars: 100 μm. n = 6. d Top panel: representative image of H&E staining in diaphragm, quadriceps femoris, tibialis anterior, triceps brachii and gastrocnemius, central and peripheral nuclei are denoted by black arrows. Bottom panel: statistical results of the number of central nuclei in diaphragm (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [0.8750, 1.0000], median difference = 0.4853, 95% CI for median difference = [0.4020, 0.5436]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.2000, rg = 0.6250, 95% CI for rg = [–0.1250, 0.8750], median difference = 0.0620, 95% CI for median difference = [–0.0160, 0.1470]), quadriceps femoris (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4615, 95% CI for median difference = [0.4400, 0.5840]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0625, 95% CI for median difference = [0.0240, 0.2130]), tibialis anterior (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4690, 95% CI for median difference = [0.4470, 0.5620]; mdx + PBS vs. mdx+MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.1000, 95% CI for median difference = [0.0150, 0.2160]), triceps brachii (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4865, 95% CI for median difference = [0.4570, 0.5120]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0705, 95% CI for median difference = [0.0050, 0.1320]), and gastrocnemius (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4700, 95% CI for median difference = [0.4410, 0.5690]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 7.5, exact p = 0.9714, rg = 0.0625, 95% CI for rg = [–0.8750, 1.0000], median difference = 0.0100, 95% CI for median difference = [–0.0200, 0.1000]). Scale bars: 100 μm. n = 4. e Quantification of body weight. WT + PBS (14.85 ± 0.14) vs. mdx + PBS (14.87 ± 0.53): t = 0.0743, p = 0.9429, Cohen’s d = 0.04 (95% CI [− 0.87, 0.96]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (14.43 ± 2.242): t = 0.4608, p = 0.6624, Cohen’s d = 0.27 (95% CI [− 0.65, 1.19]), n = 6. f Quantification of relative four-limbs grip strength. WT + PBS (8.57 ± 0.23) vs. mdx + PBS (4.90 ± 0.23): Welch’s t = 11.28, p < 0.0001, Cohen’s d = 6.51 (95% CI [4.02, 9.04]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (6.60 ± 1.29): Welch’s t = 4.55, p = 0.0012, Cohen’s d = 2.63 (95% CI [1.22, 4.05]). n = 6. g Quantification of running time to exhaustion. WT + PBS (36.66 ± 1.04) vs. mdx + PBS (20.51 ± 0.80): Welch’s t = 11.65, p < 0.0001, Cohen’s d = 6.72 (95% CI [4.15, 9.33]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (23.41 ± 0.58): Welch’s t = 4.27, p = 0.0017, Cohen’s d = 2.46 (95% CI [1.14, 3.79]). n = 6. h Quantification of total running distance to exhaustion. WT + PBS (859.70 ± 28.34) vs. mdx + PBS mice (326.90 ± 18.69): Welch’s t = 15.70, p < 0.0001, Cohen’s d = 9.06 (95% CI [5.79, 12.23]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (450.30 ± 22.21): Welch’s t = 4.252, p = 0.0018, Cohen’s d = 2.46 (95% CI [1.13, 3.79]). n = 6 . Data presented as mean ± SD. WT, wild type. GAS, gastrocnemius MyoAAV 2 A–micro-utrophin resulted in abundant vector genomes and improved muscle morphology and function. a Quantification of vector genome per diploid genome in various muscles, liver and other tissue of mdx mice injected with MyoAAV 2 A–micro-utrophin, n = 6. b Quantification of serum CK levels. WT + PBS (185.80 ± 10.20) vs. mdx + PBS (4517.00 ± 342.00): Welch’s t = 12.66, p < 0.0001, Cohen’s d = 7.31 (95% CI [4.10, 10.97]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (3417 ± 175.90): Welch’s t = 2.860, p = 0.0227, Cohen’s d = 1.65 (95% CI [0.34, 2.96]). n = 6. c Representative image (left) and quantification (right) of Evans Blue Dye (EBD) and Laminin staining of gastrocnemius from WT mice treated with PBS and mdx mice treated with MyoAAV 2A–micro-utrophin or the same volume of PBS (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0022, rg = 1.000, 95% CI for rg = [0.8750, 1.0000], median difference = 10.6698, 95% CI for median difference = [7.2296, 17.9100]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0022, rg = 1.000, 95% CI for rg = [0.8750, 1.0000], median difference = 5.2267, 95% CI for median difference = [0.3954, 15.4266]). Scale bars: 100 μm. n = 6. d Top panel: representative image of H&E staining in diaphragm, quadriceps femoris, tibialis anterior, triceps brachii and gastrocnemius, central and peripheral nuclei are denoted by black arrows. Bottom panel: statistical results of the number of central nuclei in diaphragm (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [0.8750, 1.0000], median difference = 0.4853, 95% CI for median difference = [0.4020, 0.5436]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 3, exact p = 0.2000, rg = 0.6250, 95% CI for rg = [–0.1250, 0.8750], median difference = 0.0620, 95% CI for median difference = [–0.0160, 0.1470]), quadriceps femoris (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4615, 95% CI for median difference = [0.4400, 0.5840]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0625, 95% CI for median difference = [0.0240, 0.2130]), tibialis anterior (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4690, 95% CI for median difference = [0.4470, 0.5620]; mdx + PBS vs. mdx+MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.1000, 95% CI for median difference = [0.0150, 0.2160]), triceps brachii (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4865, 95% CI for median difference = [0.4570, 0.5120]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.0705, 95% CI for median difference = [0.0050, 0.1320]), and gastrocnemius (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.0000, 95% CI for rg = [–0.5556, 1.0000], median difference = 0.4700, 95% CI for median difference = [0.4410, 0.5690]; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin: U = 7.5, exact p = 0.9714, rg = 0.0625, 95% CI for rg = [–0.8750, 1.0000], median difference = 0.0100, 95% CI for median difference = [–0.0200, 0.1000]). Scale bars: 100 μm. n = 4. e Quantification of body weight. WT + PBS (14.85 ± 0.14) vs. mdx + PBS (14.87 ± 0.53): t = 0.0743, p = 0.9429, Cohen’s d = 0.04 (95% CI [− 0.87, 0.96]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (14.43 ± 2.242): t = 0.4608, p = 0.6624, Cohen’s d = 0.27 (95% CI [− 0.65, 1.19]), n = 6. f Quantification of relative four-limbs grip strength. WT + PBS (8.57 ± 0.23) vs. mdx + PBS (4.90 ± 0.23): Welch’s t = 11.28, p < 0.0001, Cohen’s d = 6.51 (95% CI [4.02, 9.04]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (6.60 ± 1.29): Welch’s t = 4.55, p = 0.0012, Cohen’s d = 2.63 (95% CI [1.22, 4.05]). n = 6. g Quantification of running time to exhaustion. WT + PBS (36.66 ± 1.04) vs. mdx + PBS (20.51 ± 0.80): Welch’s t = 11.65, p < 0.0001, Cohen’s d = 6.72 (95% CI [4.15, 9.33]). mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (23.41 ± 0.58): Welch’s t = 4.27, p = 0.0017, Cohen’s d = 2.46 (95% CI [1.14, 3.79]). n = 6. h Quantification of total running distance to exhaustion. WT + PBS (859.70 ± 28.34) vs. mdx + PBS mice (326.90 ± 18.69): Welch’s t = 15.70, p < 0.0001, Cohen’s d = 9.06 (95% CI [5.79, 12.23]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (450.30 ± 22.21): Welch’s t = 4.252, p = 0.0018, Cohen’s d = 2.46 (95% CI [1.13, 3.79]). n = 6 . Data presented as mean ± SD. WT, wild type. GAS, gastrocnemius Fig. 5MyoAAV 2A–micro-utrophin resulted in reduced systemic inflammation and tissue fibrosis. a Quantification of serum cTnI levels. WT + PBS (16.56 ± 6.43) vs. mdx + PBS (16.56 ± 4.54): Welch’s t = 0, p > 0.9999, Cohen’s d = 0. mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (15.26 ± 6.71): Welch’s t = 0.393, p = 0.7035, Cohen’s d = 0.23 (95% CI [-0.91, 1.36]). n = 6. b Quantification of serum NT-proBNP levels. WT + PBS (66.54 ± 25.41) vs. mdx + PBS (79.56 ± 28.10): Welch’s t = 0.842, p = 0.4196, Cohen’s d = 0.49 (95% CI [-0.66, 1.64]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (72.48 ± 29.39): Welch’s t = 0.427, p = 0.6788, Cohen’s d = 0.25 (95% CI [-0.90, 1.40]). n = 6. c Quantification of serum . ALT levels. WT + PBS (12.90 ± 3.50) vs. mdx + PBS (18.76 ± 6.68): Welch’s t = 1.903, p = 0.0952, Cohen’s d = 1.10 (95% CI [-0.19, 2.39]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (16.87 ± 7.46): Welch’s t = 0.462, p = 0.6528, Cohen’s d = 0.27 (95% CI [-0.86, 1.40]). n = 6. d Quantification of serum AST levels. WT + PBS (166.61 ± 60.77) vs. mdx + PBS (222.33 ± 61.65): Welch’s t = 1.577, p = 0.1495, Cohen’s d = 0.91 (95% CI [-0.28, 2.10]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (188.90 ± 88.41): Welch’s t = 0.760, p = 0.4669, Cohen’s d = 0.44 (95% CI [-0.71, 1.58]). n = 6. e Quantification of serum IL-6 levels. WT + PBS (2.53 ± 5.21) vs. mdx + PBS (58.47 ± 24.19): Welch’s t = 5.538, p = 0.0020, Cohen’s d = 3.20 (95% CI [1.49, 4.91]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (32.06 ± 13.78): Welch’s t = 2.324, p = 0.0489, Cohen’s d = 1.34 (95% CI [0.09, 2.59]). f Quantification of serum C5a levels. WT + PBS (17.84 ± 3.00) vs. mdx + PBS (93.31 ± 24.53): Welch’s t = 7.48, p = 0.0006, Cohen’s d = 4.32 (95% CI [1.63, 7.01]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (60.44 ± 21.13): Welch’s t = 2.49, p = 0.0326, Cohen’s d = 1.44 (95% CI [0.04, 2.83]). n = 6. g Representative image (left) and quantification (Rright) of Masson staining of myocardium from WT mice treated with PBS and mdx mice treated with MyoAAV 2 A–micro-utrophin or the same volume of PBS. mdx + PBS compared to mdx + MyoAAV 2 A–micro-utrophin (U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0283, 95% CI for median difference = [0.0220, 0.0340]). Scale bars: 500 μm. n = 4. h Left panel: representative image of H&E staining in myocardium and diaphragm, inflammatory infiltration is indicated by white arrow. Right panel: statistical results of the inflammatory infiltration area of diaphragm (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0283, 95% CI for median difference = [0.0220, 0.0340]; mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0113, 95% CI for median difference = [0.0020, 0.0207]). Scale bars: 500 μm. n = 4 MyoAAV 2A–micro-utrophin resulted in reduced systemic inflammation and tissue fibrosis. a Quantification of serum cTnI levels. WT + PBS (16.56 ± 6.43) vs. mdx + PBS (16.56 ± 4.54): Welch’s t = 0, p > 0.9999, Cohen’s d = 0. mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin (15.26 ± 6.71): Welch’s t = 0.393, p = 0.7035, Cohen’s d = 0.23 (95% CI [-0.91, 1.36]). n = 6. b Quantification of serum NT-proBNP levels. WT + PBS (66.54 ± 25.41) vs. mdx + PBS (79.56 ± 28.10): Welch’s t = 0.842, p = 0.4196, Cohen’s d = 0.49 (95% CI [-0.66, 1.64]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (72.48 ± 29.39): Welch’s t = 0.427, p = 0.6788, Cohen’s d = 0.25 (95% CI [-0.90, 1.40]). n = 6. c Quantification of serum . ALT levels. WT + PBS (12.90 ± 3.50) vs. mdx + PBS (18.76 ± 6.68): Welch’s t = 1.903, p = 0.0952, Cohen’s d = 1.10 (95% CI [-0.19, 2.39]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (16.87 ± 7.46): Welch’s t = 0.462, p = 0.6528, Cohen’s d = 0.27 (95% CI [-0.86, 1.40]). n = 6. d Quantification of serum AST levels. WT + PBS (166.61 ± 60.77) vs. mdx + PBS (222.33 ± 61.65): Welch’s t = 1.577, p = 0.1495, Cohen’s d = 0.91 (95% CI [-0.28, 2.10]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (188.90 ± 88.41): Welch’s t = 0.760, p = 0.4669, Cohen’s d = 0.44 (95% CI [-0.71, 1.58]). n = 6. e Quantification of serum IL-6 levels. WT + PBS (2.53 ± 5.21) vs. mdx + PBS (58.47 ± 24.19): Welch’s t = 5.538, p = 0.0020, Cohen’s d = 3.20 (95% CI [1.49, 4.91]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (32.06 ± 13.78): Welch’s t = 2.324, p = 0.0489, Cohen’s d = 1.34 (95% CI [0.09, 2.59]). f Quantification of serum C5a levels. WT + PBS (17.84 ± 3.00) vs. mdx + PBS (93.31 ± 24.53): Welch’s t = 7.48, p = 0.0006, Cohen’s d = 4.32 (95% CI [1.63, 7.01]). mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin (60.44 ± 21.13): Welch’s t = 2.49, p = 0.0326, Cohen’s d = 1.44 (95% CI [0.04, 2.83]). n = 6. g Representative image (left) and quantification (Rright) of Masson staining of myocardium from WT mice treated with PBS and mdx mice treated with MyoAAV 2 A–micro-utrophin or the same volume of PBS. mdx + PBS compared to mdx + MyoAAV 2 A–micro-utrophin (U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0283, 95% CI for median difference = [0.0220, 0.0340]). Scale bars: 500 μm. n = 4. h Left panel: representative image of H&E staining in myocardium and diaphragm, inflammatory infiltration is indicated by white arrow. Right panel: statistical results of the inflammatory infiltration area of diaphragm (WT + PBS vs. mdx + PBS: U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0283, 95% CI for median difference = [0.0220, 0.0340]; mdx + PBS vs. mdx + MyoAAV 2 A–micro-utrophin: U = 0, exact p = 0.0286, rg = 1.000, 95% CI for rg = [0.5000, 1.0000], median difference = 0.0113, 95% CI for median difference = [0.0020, 0.0207]). Scale bars: 500 μm. n = 4 Rationally designed micro-utrophin delivered via MyoAAV 2 A exhibited a favorable safety profile in mdx mice No significant differences in serum cTnI levels were observed between WT + PBS and mdx + PBS mice (t = 0, p > 0.9999; Fig. 5a) or between mdx + PBS and mdx + MyoAAV 2A–micro-utrophin mice (t = 0.393, p = 0.7035; Fig. 5a). Similarly, NT‑proBNP levels did not differ between WT + PBS and mdx + PBS (t = 0.842, p = 0.4196) or between mdx + PBS and mdx + MyoAAV 2A–micro-utrophin mice (t = 0.427, p = 0.6788; Fig. 5b). ALT levels showed no significant differences between WT + PBS and mdx + PBS (t = 1.903, p = 0.0952) or between mdx + PBS and mdx + MyoAAV 2A–micro-utrophin mice (t = 0.462, p = 0.6528; Fig. 5c). AST levels also showed no significant differences (WT + PBS vs. mdx + PBS: t = 1.577, p = 0.1495; mdx + PBS vs. mdx + MyoAAV 2A–micro-utrophin mice: t = 0.760, p = 0.4669; Fig. 5d). In contrast, serum IL‑6 levels were significantly elevated in mdx + PBS mice compared to WT + PBS (t = 5.538, p = 0.0020) and were significantly reduced after treatment (t = 2.324, p = 0.0489; Fig. 5e). Serum C5a levels were also markedly higher in mdx + PBS than in WT + PBS (t = 7.48, p = 0.0006) and were significantly decreased by treatment (t = 2.49, p = 0.0326; Fig. 5f). Moreover, Masson’s trichrome staining revealed that MyoAAV 2A–micro-utrophin treated mdx mice exhibited reduced fibrosis area in the myocardium (U = 0, exact p = 0.0286, rg = 1.000; Fig. 5g). H&E staining indicated that systemic administration of MyoAAV 2A–micro-utrophin also significantly alleviated the inflammatory infiltration area in the diaphragm (U = 0, exact p = 0.0286, rg = 1.000; Fig. 5h) and had no toxic effect in the myocardium (Fig. 5h) or liver (Supplementary Fig. 5), as no inflammatory infiltration area was observed among the three groups. Discussion The assessment of safety and efficacy remains paramount in developing gene therapy vectors. While rAAV has emerged as the predominant delivery platform for genetic interventions due to its favorable tropism and persistent expression characteristics, critical limitations persist in clinical applications. Dose-dependent toxicity represents a particularly pressing concern, as elevated vector doses intended to enhance therapeutic efficacy may paradoxically exacerbate immune-mediated adverse events (AEs). The risk-benefit dichotomy is exemplified by recent clinical outcomes in X-linked myotubular myopathy therapy. The AAV8-mediated MTM1 gene replacement trial demonstrated functional improvements across multiple endpoints, but tragically resulted in three fatalities in the high-dose cohort (3 × 1014 vg/kg). These patients succumbed to sepsis and subsequent gastrointestinal hemorrhage within 4–6 weeks post-administration, highlighting the critical threshold of vector tolerance [28]. Similarly, phase 3 EMBARK trial data further corroborate these safety concerns, revealing significant treatment-related AEs in DMD patients aged ≥ 4 years receiving high-dose ELEVIDYS administration [29]. These collective findings underscore a dose-dependent safety paradox in rAAV therapeutics, while higher viral loads may enhance transduction efficiency, they simultaneously elevate risks of innate immune activation and cellular immune responses to the vector [6, 30]. This mechanistic conundrum necessitates urgent development of next-generation delivery platforms. Potential solutions may include engineered capsids with enhanced tropism, immunosuppressive adjuvant therapies, or innovative dose fractionation strategies to mitigate toxicity while maintaining therapeutic efficacy. MyoAAV, a muscle-targeting engineered capsid, demonstrates remarkable tropism for skeletal and cardiac muscle, enabling efficient gene delivery at substantially lower doses. In our study, MyoAAV-mediated delivery of micro-utrophin genes results in robust muscle-specific expression with minimal off-target transduction in the liver (Supplementary Fig. 3), thereby reducing the risks of hepatotoxicity and innate immune activation, common limitations of systemic rAAV administration [3]. Furthermore, utrophin, an autosomal paralog of dystrophin, shares significant structural and functional similarity and has been consistently shown to compensate for dystrophin deficiency in multiple preclinical models [22, 23, 31]. Crucially, as a self-protein endogenously expressed in patients, utrophin poses a lower risk of immune recognition compared to dystrophin, which is often perceived as foreign in DMD patients due to its congenital absence. Previous studies, including ours, have demonstrated that systemic delivery of AAV-micro-utrophin prevented myonecrosis without eliciting immunogenicity in murine and canine DMD models [23, 24, 32]. These attributes make utrophin a highly promising therapeutic candidate. However, the substantial size of utrophin (395 kDa) exceeds the packaging capacity of AAV vectors, necessitating the development of truncated yet functional micro-utrophin variants that retain critical domains. In this study, we report that a rationally engineered micro-utrophin (ΔR4–R22/ΔC + H3/H4), delivered using the muscle-tropic MyoAAV2A capsid, confers therapeutic efficacy in a mouse model of DMD. ΔR4-R22/ΔC + H3/H4 retains the NT, the first three SRs, a novel hinge region and CR domain. The NT and CR domain mediate binding to F-actin and the DGC, respectively, forming a mechanical link between the cytoskeleton and extracellular matrix [33, 34]. The retention of the first three SRs is supported by recent evidence indicating their high lipid-binding affinity, which is essential for sarcolemmal association and force transmission [35]. While the C-terminal segment of hinge 4 was retained for its connection to the CR domain, mirroring full-length utrophin, hinge 2 was replaced with hinge 3 owing to its structural vulnerability in prior constructs. The C-terminal domain was omitted to facilitate efficient packaging into MyoAAV2A, consistent with other miniaturized utrophin/dystrophin designs [27, 36]. Thus, our micro-utrophin (ΔR4–R22/ΔC + H3/H4) differs from previously reported constructs (ΔR4–R21/ΔCT) [24, 27, 37, 38], not only by the exclusion of R22 but also through the incorporation of a novel hybrid hinge (H3/H4) linking the SRs to the CR domain. Following systemic delivery at a low dose of 1 × 1012 vg/kg, robust micro-utrophin expression was confirmed via western blotting and immunofluorescence, with pronounced muscle-specific transduction and minimal liver exposure, highlighting the translational advantage of MyoAAV for neuromuscular disorders. Importantly, expression was observed not only in limb muscles but also in the diaphragm and heart, tissues critically involved in DMD-related mortality. Although tissue-specific heterogeneity in expression levels was consistent with previous reports [36], these findings support the potential of this engineered micro-utrophin to address respiratory and cardiac dysfunction in DMD, pending further functional validation [39, 40]. Additionally, systemic delivery of the MyoAAV 2A–driven micro-utrophin induced significant, differential recruitment of sarcoglycan complexes across nearly all striated muscles examined, with γ-sg recruitment being notably stronger than that of α-sg, consistent with a prior study [36]. The compartmentalized regulatory pattern may arise from (1) differential anchoring mechanisms of α-sg and γ-sg in the cell membrane and their interactions with utrophin [25], (2) tissue-specific post-translational modification of the micro-utrophin–sarcoglycan complex [41, 42], or (3) variations in endogenous utrophin compensation mechanisms between embryonic-derived (e.g., diaphragm) and somite-derived muscles [43]. Of note, there is a discrepancy between western blotting and immunofluorescence analyses for γ-sg and α-sg expression in mdx mice which was partly attributed to the insensitive indicator that western blotting measures total protein levels and may not accurately reflect the functional recruitment of the sarcoglycan complex. Immunofluorescence staining for γ- and α-sg localization is more relevant to functional recovery. Furthermore, this study identified a pivotal issue: the successful delivery of micro-utrophin failed to induce the downregulation of endogenous full-length utrophin as detected by anti-Utro-C antibody in Fig. 3a. This observation aligns with a previous report, which found that endogenous utrophin remained detectable at 3 months post-treatment following utrophin-based therapy [44]. At only 4 weeks post-treatment, although micro-utrophin expression was successful, the muscle may not have undergone complete functional remodeling and the signaling pathways triggering utrophin downregulation may not yet be fully activated. Future long-term studies will be essential to monitor the dynamic changes in endogenous utrophin over an extended treatment period and to elucidate its regulatory mechanisms using advanced detection methods. The successful expression of micro-utrophin delivered by MyoAAV 2A at the low dose of 1.0 × 10¹² vg/kg significantly restored the sarcolemmal localization of γ-sg and α-sg, and ameliorated muscle pathology in mdx mice. This was accompanied by significant functional improvement, as evidenced by the recovery of grip strength to 86% of wild-type levels and a reduction in serum CK to 76% of untreated mdx levels, observed as early as 4 weeks post-injection. These initial findings show promising efficacy when contextualized with prior studies. For example, a previous mini-utrophin construct (ΔR4–R21/ΔCT) delivered via AAV6 at 1.2 × 1014 vg/kg restored grip strength to 93% of wild-type levels and reduced serum creatine kinase (CK) to 60% of untreated mdx levels at 12 weeks post-treatment [37]. Similarly, a study from our group using AAV9 to deliver a micro-utrophin construct at a high dose (~ 1 × 1014 vg/kg) also reported functional recovery approaching wild-type levels as early as 8 weeks post-injection [24]. The robust micro-utrophin expression and sarcolemmal DGC restoration observed in our study at just 4 weeks suggest a rapid onset of action. While direct comparisons are constrained by differences in study design and transgenes, the compelling efficacy achieved here with a significantly lower vector dose highlights the potential of our novel micro-utrophin design combined with MyoAAV 2A. Moreover, systemically administered MyoAAV 2 A delivered micro‑utrophin at 1 × 10¹² vg/kg exhibited a favorable safety profile in mdx mice. No significant differences in cardiac (cTnI, NT‑proBNP) or hepatic (ALT, AST) biomarkers were observed between treated and control groups, indicating absence of cardiotoxicity or hepatotoxicity. Importantly, serum IL‑6 and C5a levels—elevated in untreated mdx mice due to chronic inflammation—were significantly reduced after treatment, suggesting that micro‑utrophin ameliorates systemic inflammation without triggering immune‑mediated adverse events. Histological analyses further supported safety as treated mice showed reduced myocardial fibrosis and alleviated diaphragmatic inflammation, with no toxic effects in the myocardium or liver. Collectively, these findings demonstrate that combining a muscle‑tropic MyoAAV 2 A capsid with a rationally truncated micro‑utrophin at a relatively low dose may be a promising strategy to circumvent dose‑dependent toxicity while preserving therapeutic benefits. However, some features of this study require caution in the interpretation of the findings. First, the therapeutic efficacy and safety were evaluated using a neonatal mdx mouse model with intraperitoneal injection. Neonatal administration may overestimate transduction efficiency due to immune immaturity and heightened vector permissiveness in developing tissues. In clinical practice, DMD patients are typically diagnosed and treated after 4 years of age, when the immune system is fully mature. Therefore, the results obtained in neonatal mice may not directly translate to juvenile or adult settings. Future studies should validate the therapeutic potential of this approach in older mdx mice (e.g., 4–6 weeks or 3‑month‑old). Second, the observation period was limited to 4 weeks post‑injection. Long‑term efficacy and safety beyond this time point remain unknown; thus, extended studies lasting 3–6 months are necessary to evaluate durability of transgene expression and potential late‑onset adverse effects. Third, we did not assess humoral or cellular immune responses against the vector (neutralizing antibodies), the micro‑utrophin transgene product, or capsid‑specific T cell responses. Additionally, the transgene was driven by a CMV promoter, which may be subject to silencing over time. The use of muscle‑specific promoters (e.g., MHCK7, CK8) could potentially enhance long‑term stability and reduce immunogenicity, and this warrants investigation in future iterations of the construct. In conclusion, the favorable results obtained within our abbreviated detection window demonstrate that our rationally designed micro-utrophin construct, when delivered via the muscle-directed MyoAAV 2 A capsid, can achieve rapid and significant therapeutic effects under conditions of low immune response. This approach holds promise for mitigating the risks associated with high-dose AAV therapy by enabling robust efficacy at substantially lower doses, thereby potentially reducing capsid- and transgene-related toxicities. While the present study establishes proof-of-concept in a neonatal screening model, we recognize the importance of evaluating this strategy under clinically relevant conditions. A formal dose-ranging study in our follow-up investigation, with considerable room for a 1-log (10-fold) or greater dose reduction, may help to further explore the translational potential of this approach. Additional studies, particularly in large animal models using systemic intravenous delivery, would be necessary to establish safety and efficacy profiles before any clinical development can be considered. Collectively, our findings suggest that, with further validation, combining an optimized micro-utrophin construct with a highly muscle-tropic AAV capsid could represent a promising strategy to achieve safer and potentially more effective treatments for DMD. Materials and methods Micro-utrophin transgene design and vector production Based on bioinformatics and phylogenomic analysis of sequence conservation described previously [24], we engineered a synthetic micro-utrophin transgene (ΔR4–R22/ΔC + H3/H4) containing the NT, the first three SRs, a novel hinge region, and the CR domain. A hybrid hinge comprising the N-terminal segment of hinge 3 and the C-terminal segment of hinge 4 was designed to link the SRs to the CR domain. The transgene was derived from wild-type utrophin and codon-optimized using the most frequently used codons in myosin, the most highly expressed protein in muscle to enhance expression efficiency. The synthesized micro-utrophin cDNA was subsequently subcloned into a muscle-specific MyoAAV 2A expression vector, which was constructed as previously reported [13]. The construct employs the human cytomegalovirus (CMV) promoter followed by a CW3SL regulatory element [45]. The MyoAAV 2A–CMV-micro-utrophin-CW3SL was generated and purified by the Innovation Academy for Precision Measurement Science and Technology (CAS) vector core using the triple-plasmid transfection method in HEK 293T cells [46]. Murine model vector administration C57BL/10SnJ (wild type) and mdx mice were purchased from the Jackson Laboratory. This study utilized 16 C57BL/10 mice (9 male and 7 female) and 36 mdx mice (20 male and 16 female), all injected at 3 days of age. Prior to receiving intraperitoneal injection of MyoAAV 2 A–micro-utrophin or phosphate-buffered saline (PBS), individual pups were toe-tattooed with the Aramis Micro tattoo kit (Ketchum Manufacturing Inc., Brockville, ON, Canada) and randomly assigned to different treatment groups. All injections and subsequent tissue collection were performed by investigators blinded to the experimental conditions. According to the experimental protocol, wild-type and mdx pups received 25–50 µl of either PBS (negative control) or MyoAAV 2 A–micro-utrophin diluted in PBS via a 32-gauge insulin syringe. Each mouse was weighed immediately before injection. Following vector administration, all animals were returned to their original litters and separated after weaning. Murine model tissue procurement and storage At approximately four weeks of age, mdx and wild type mice were euthanized with isoflurane in compliance with institutional guidelines. The tibialis anterior, gastrocnemius, quadriceps, triceps, diaphragm, heart, lung, spleen, kidney, gonad and liver were collected for further processing. Designated tissue samples for histological analysis were immediately fixed in 10% neutral-buffered formalin, followed by conventional processing including dehydration (ASP200S, Leica Biosystems, Solms, Germany) and paraffin embedding. Sections were cut at a thickness of 8 μm. Additional tissue samples designated for molecular analysis were placed in cryovials, rapidly frozen in liquid nitrogen, and stored at −80 °C. Immunoblot analysis Designated biological tissue samples were lysed in lysis buffer (89901, ThermoFisher, Rockford, IL, USA) supplemented with phosphatase inhibitors (5892970001, Roche Diagnostics, Mannheim, Germany), and proteinase inhibitors (11697498001, Roche Diagnostics) for 30 min on ice. The lysates were then centrifuged at 13,000 × g for 10 min at 4 °C (5424R, Eppendorf SE, Hamburg, Germany) to remove insoluble debris. Protein samples (20–40 µg per lane) were separated by SDS-PAGE and transferred to PVDF membranes (MISEQ00010, Millipore, Burlington, MA, USA). The membranes were activated by baking at 37 °C for 30 min, followed by brief immersion in methanol for 3 s and rinsing in 1× PBS for 5 min. After blocking with 5% non-fat dry milk for 1 h at room temperature, the membranes were incubated with primary antibodies overnight at 4 °C. The following primary antibodies were used: mouse polyclonal anti-micro-utrophin (1:200, sc-33699, Santa Cruz Biotechnology, Inc., Dallas, TX, USA) targeting the N-terminal epitope; rabbit polyclonal anti-γ-sarcoglycan (1:300, NBP1-59744, Novus Biologicals, Centennial, CO, USA); rabbit monoclonal anti-α-sarcoglycan (1:2000, ab189254, Abcam, Cambridge, UK); and mouse polyclonal anti-vinculin (1:10000, V9131, Merck KGaA, Darmstadt, Germany) as a loading control. After three washes in 0.1% TBST, membranes were incubated for 1 h at room temperature with IRDye 680LT/800CW-conjugated secondary antibodies (anti-mouse IgG, 926-68020/926-32210; anti-rabbit IgG, 926-68021/926-32211; all at 1:10,000 dilution, LI-COR, Inc., Nebraska, NE, USA). Following additional washes, images were acquired using an Odyssey Imaging System (Odyssey DLx, LI-CO Biosciences, Lincoln, NE, USA). To ensure quantification reliability, exposure times were optimized for each blot to avoid signal saturation. Images were acquired only when signal intensities remained below the saturation threshold, as monitored by the imaging software. All quantified signals were confirmed to be within the linear dynamic range of the detection system. During image analysis using ImageJ, local background subtraction was applied to each band to correct for non-specific signals and membrane heterogeneity. For micro-utrophin quantification, the raw signal was normalized to the corresponding vinculin signal from the same lane to correct for variations in protein loading, yielding a normalized ratio (micro-utrophin/vinculin) for each sample in arbitrary units. These values were used directly for statistical comparisons between groups, but cannot be compared directly across different blots or experiments. For α-sg and γ-sg quantification, the normalized values were then expressed as fold change relative to the average signal of the control group (e.g., PBS-treated C57 mice), which was set to 1. Immunofluorescence staining Immunofluorescence staining of utrophin, α-sg and γ-sg was performed on sections of all muscle specimens. For utrophin detection, three serial sections were separately stained with three distinct antibodies: a mouse monoclonal antibody (1:50, sc-33699, Santa Cruz Biotechnology, Inc.) targeting the N-terminus to detect both full-length and micro-utrophin, a mouse monoclonal antibody (1:50, sc-33700, Santa Cruz Biotechnology, Inc.) targeting the C-terminus specific to full-length utrophin, and an anti-laminin-2 antibody (1:200, L0663, Merck KGaA ). Sections were permeabilized with 1% Triton X-100 (T8200, Solarbio, Beijing, China) in 0.01 M PBS for 20 min, followed by three 5-minute washes in PBS. After incubation with the different primary antibodies, slides were washed again, and then incubated with goat anti-mouse IgG Alexa Fluor® 594 (1:500, ab150116, Abcam), or FITC-conjugated goat anti-rat IgG (1:2000, ab6840, Abcam) for 30 min at room temperature. Finally, sections were mounted with DAPI-containing medium (C0065, Solarbio), and imaged using a Leica DM6000B microscope. The same staining procedure was applied for α-sg and γ-sg/laminin co-staining. α-sg was detected using a rabbit monoclonal antibody (1:1000, ab189254, Abcam) and goat anti-rabbit IgG Alexa Fluor® 594 (1:500, ab150080, Abcam). γ-sg was detected using a rabbit polyclonal antibody (1:300, NBP1-59744, Novus Biologicals, Centennial, CO, USA) and the same secondary antibody. It should be noted that non-specific interstitial signal, attributable to the use of murine antibodies on mouse tissue, was observed but did not interfere with the assessment of utrophin membrane localization. Quantification of vector genome copies per diploid genome Genomic DNA was extracted from mouse tissues using the Blood & Tissue Genome DNA Extract Kit (B1111B, TIANGEN, Beijing, China) according to the manufacturer’s protocol. DNA concentration was determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) by measuring absorbance at 260 nm. The vector genome copy number per diploid genome was quantified by TaqMan-based real-time quantitative PCR (qPCR) on a Real-Time PCR System (QuantStudio 7 Pro, Thermo Fisher Scientific). The qPCR reaction was performed in a 20 µL volume containing 10 µL of 2X Pro Taq HS Probe Premix III (AG11730, ACCURATE BIOTECHNOLOGY(HUNAN) Co., Ltd., Changsha, China), 0.2 µM each of forward and reverse primers, 0.4 µM TaqMan probe, and ~ 80 ng of genomic DNA template. The mouse Apob gene (apolipoprotein B; two copies per diploid genome) was used as an endogenous reference. Primers and probe for Apob were obtained from Jackson website (oIMR1544 for forward primer, oIMR3580 for reverse primer, TmolMR0105 for probe). Primers and probe targeting the micro-utrophin sequence were designed using Beacon Designer 7 software and synthesized by ACCURATE BIOTECHNOLOGY(HUNAN) Co., Ltd, sequences are provided in Supplementary Table 1. The qPCR cycling conditions were as follows: 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. All samples were run in triplicate. No-template controls were included in each run to monitor potential contamination. Amplification efficiencies for both Apob and micro-utrophin primers were validated using a 6-point 10-fold serial dilution of mouse genomic DNA (ranging from 108 copies/µL to 103 copies/µL). Standard curves were generated by plotting Ct values against the log of template amount. The amplification efficiency ranged from 98% to 102% (R² > 0.99), confirming that the efficiency assumption required for the comparative Ct method was satisfied. The vector genome copies per diploid genome were then calculated using the ΔCt method as: 2×2^(Ct_Apob – Ct_micro-utrophin). The resulting value represents the average number of AAV vector genomes per diploid genome equivalent (i.e., per mouse cell). All data are presented as mean ± SD. Hematoxylin-eosin staining Designated biological tissue samples were fixed in 10% formaldehyde solution and were continuously sliced to 8-μm thick cross-sections. Histology was examined on sections stained by hematoxylin-eosin according to the recommended process. Images were captured with a Leica microscope (Leica GT450, Leica) for demonstration of basic histology, including detection of muscle-fibers containing internal myonuclei and inflammation. Masson staining Diaphragm and myocardium were collected, fixed in 4% paraformaldehyde, and embedded in paraffin. Transverse sections (5 μm) were stained with Masson’s trichrome (G1340, Solarbio) following the manufacturer’s protocol: after deparaffinization and rehydration, sections were treated with Bouin’s solution (1 h), stained with Weigert’s iron hematoxylin (8 min), Biebrich scarlet‑acid fuchsin (5 min), differentiated in phosphomolybdic‑phosphotungstic acid (10 min), and finally stained with aniline blue (5 min) for collagen. Images were acquired using a brightfield microscope, and fibrosis area (blue) was quantified with ImageJ software as a percentage of the total tissue area. EBD treatment To assess muscle fiber membrane stability following treatment with MyoAAV 2A–micro-utrophin for 4 weeks, mice were injected intraperitoneally with EBD (314-13-6, Solarbio) at a dose of 0.1 ml/10 g mouse body weight 24 h prior to tissue sampling. One hour after EBD injection, mice were subjected to a downhill running protocol on a motorized treadmill set at a −15° decline. The running speed was maintained at 10–20 m/min for 30 min [47]. Twenty-four hours after the completion of the exercise bout, mice were euthanized and the gastrocnemius were harvested. Muscle tissues were immediately frozen in liquid nitrogen-cooled isopentane and stored at −80 °C. Cryosections (10 μm thick) were stained with laminin, and examined under a fluorescence microscope (Leica DM6000B, Leica Microsystems, Wetzlar, Germany). Fibers exhibiting red fluorescence were identified as those with compromised sarcolemmal integrity. The percentage of EBD-positive fibers relative to total fibers, or the EBD-positive area per muscle cross-section, was quantified using ImageJ software. Measurement of serum biomarkers Serum levels of cardiac troponin I (cTnI), N‑terminal pro‑brain natriuretic peptide (NT‑ProBNP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), interleukin‑6 (IL‑6), complement component C5a, and creatine kinase (CK) were measured using commercial assay kits according to the manufacturers’ protocols. cTnI, NT‑ProBNP, and ALT were quantified using ELISA kits from Jianglai Biotechnology (Shanghai, China; JL11280, JL11641, and JL12668, respectively). AST and IL‑6 were measured using the Mouse AST SimpleStep ELISA Kit (ab263882, Abcam) and the Mouse IL‑6 SimpleStep ELISA Kit (ab222503, Abcam), respectively, both following a single sandwich ELISA protocol. Serum C5a levels were determined using the Mouse Complement C5a ELISA Kit (ab193718, Abcam); briefly, 100 µL of standards or diluted serum samples were added to pre‑coated plates and incubated for 2.5 h, followed by biotinylated detection antibody (1 h), HRP‑streptavidin (45 min), TMB substrate (30 min in the dark), and stop solution, with absorbance read at 450 nm. CK activity was measured using the Creatine Kinase Activity Assay Kit (ab155901, Abcam) by monitoring absorbance at 450 nm kinetically for 10–40 min at 37 °C. All samples were run in duplicate, and concentrations were calculated by interpolation from the respective standard curves. Grip strength measurement Mouse grip strength was assessed with a digital grip strength meter (YLS-13 A, Ji-Nan Biotechnology, Jinan, China) according to a standardized protocol [48]. Before testing, animals were acclimated to the experimental environment for 10 min. Each mouse was then gently placed to grip the stainless steel pull bar with all four limbs. A trained experimenter pulled the mouse backward horizontally by the tail at a consistent speed until the grip was released. The maximum force (in grams) at the point of detachment was automatically recorded by the device [49]. Each animal underwent five consecutive trials with 40-second rest intervals between trials to avoid fatigue. The final grip strength value was calculated as the average of the three highest measurements from valid trials. Treadmill exercise Exercise capacitywas measured using an adapted protocol described in a published paper [50]. The treadmill (DSPT-216; Hangzhou Duan’s Manufacturing factory, Hangzhou, China) was set to a 0% slope. All mice underwent a 3-day acclimation period (5 min/day at 5 m/min, no shock) before the formal test to minimize stress and inter-individual motivational differences. Following acclimation, the speed was increased by 1 m/min each minute until exhaustion. Exhaustion was defined as the point at which the mouse could no longer maintain running for 10 s despite mechanical stimulation. To address potential motivational bias, a low-intensity electric shock (0.2 mA) was delivered through a grid placed at the rear of the treadmill lane only when the mouse remained on the grid for > 3 consecutive seconds; this served solely as a mild motivational stimulus. During the test, an investigator blinded to group assignment manually encouraged any mouse that stopped running for > 5 s by gentle tapping; if the mouse failed to resume running within 10 s, the test was terminated, and the last completed speed and distance were recorded. Blinding was confirmed by an independent observer who verified that the experimenter had no access to group codes until after data analysis. Statistics For n = 3–4/group, normality could not be reliably assessed, the Mann-Whitney U test (exact two-tailed) was used for two pre-specified comparisons (C57 + PBS vs. mdx + PBS; mdx + PBS vs. mdx + MyoAAV–micro-utrophin). Effect sizes reported as Glass rank biserial correlation(rg) and their 95% CIs were obtained by bootstrap (5,000 resamples); median differences and 95% CIs were estimated via Hodges-Lehmann. For n = 6/group, normality was first tested with Shapiro-Wilk. If normality was met (p > 0.05), Welch t-test was applied, and effect sizes are reported as Cohen’s d with 95% confidence intervals calculated using the noncentral t‑distribution method, otherwise, Mann-Whitney U test was used. Analyses were performed using R 4.2 (package rcompanion) and Prism 10.6.1. For all comparisons, two-tailed p-value < 0.05 was considered statistically significant. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3 Supplementary Material 4 Supplementary Material 4 Supplementary Material 5 Supplementary Material 5 Supplementary Material 6 Supplementary Material 6 Supplementary Material 7 Supplementary Material 7 Supplementary Material 8 Supplementary Material 8