Fig 2: Dmd and Utrn expression during mouse ontogeny.a RT-qPCR analysis of Dmd (upper panel) and Utrn (lower panel) mRNA expression in limb tissue during mouse development. Sampled as highlighted in grey. Expression levels were normalised to 18 S and Tbp. Bars show mean + /- SEM; comparison between ages was performed using a one-way ANOVA (p < 0.0001 for Dmd and p = 0.008 for Utrn) with Tukey’s multiple comparisons test (adjusted p values are shown: *p < 0.05, ***p < 0.001, ****p < 0.0001). b, c Representative bioluminescent images (b) and quantification (c) of limb radiance from UtrnR+/- embryos sampled during embryogenesis. d Spectral unmixing of DmdG+/-, UtrnR+/- and DmdG+/- / UtrnR+/- E13.5 embryos. The green component (representing CBG99Luc) was only detected in DmdG+/- and DmdG+/-/UtrnR+/- embryos, while the red component (representing RFluc) was detected in UtrnR+/- and DmdG+/-/UtrnR+/- embryos. e X-Gal staining of lacZ expression in UtrnR+/- E13.5 embryo, visualised by optical projection tomograph (OPT). LacZ signal is shown in red with total body fluorescence, used to visualise mass, shown in grey. LacZ signal was strongest in the developing digits and spine.

Fig 3: Imaging Dmd and Utrn expression in vivo using bioluminescence.a, b Schematic of the knock in (KI) strategy used to generate DmdG (a) and UtrnR (b) reporter mouse lines. Reporter genes were inserted into the final exon of the endogenous genes, between the stop codon and the 3’UTR. a The CBG99Luc gene was knocked into endogenous Dmd, separated from the coding region of exon 79 with a T2A site. b LacZ and RFluc genes were knocked into endogenous Utrn, separated by T2A sites. c Bioluminescent imaging of adult (P28) heterozygous DmdG+/- female mice showed strong bioluminescent signal corresponding with the location of the musculature and head, while no signal was visible in a wild type (WT) control mouse (DmdG-/-). d Quantification of whole-body radiance emitted from DmdG+/+, DmdG+/-, and WT mice showed that bioluminescence emitted from homozygous (DmdG+/+) females was around double that emitted by hemizygous (DmdG+/-) females and equivalent to DmdG+/y males. Significantly more bioluminescence was detected in all DmdG+ mice than in WT animals. Bar graphs show mean + /- SEM. For each sex, comparisons between genotypes were performed using one-way ANOVA (p < 0.0001) with Sidak’s multiple comparisons test (adjusted p values: **p < 0.01, ***p < 0.001, ****p < 0.0001). e Ex vivo bioluminescent imaging of tissues dissected from heterozygous (DmdG+/-) female mice; highest levels of radiance (blue green) were observed in the skeletal muscle (SkM) and heart, with lower signal in white adipose (WAT) and kidney. Quantification (below) shows mean radiance (p/s/cm2/sr) (n = 3 + /- SD). f Allele-specific RT-qPCR, using primers that distinguish mRNA derived from the KI allele (DmdG+, white bars) and the WT allele (DmdG-, grey bars); comparable expression of both alleles was observed in tissues from adult DmdG+/- female mice. g Bioluminescence in adult (P28) heterozygous (UtrnR+/-) female mice was detected particularly across the abdomen and head, but not in the WT (UtrnR-/-) animal. h Quantification of whole-body radiance emitted from UtrnR+/+, UtrnR+/- and WT mice. For both sexes, homozygous (UtrnR+/+) mice emitted approximately twice the flux of heterozygous (UtrnR+/-) mice. Radiance was significantly higher in all UtrnR+ mice than WT mice (UtrnR-/-). Statistical analysis performed as of d. i Bioluminescence imaging of ex vivo tissues from UtrnR+/- mice showed strong signal in WAT and kidney, but less in the SkM and heart. Quantification (below) shows mean radiance (p/s/cm2/sr) (n = 3 + /- SD). j Allele-specific RT-qPCR on tissues dissected from heterozygous (UtrnR+/-) mice showed comparable mRNA expression from the targeted allele (UtrnR+, white bars) and the wildtype allele (UtrnR-, grey bars). k Expression of lacZ in tissues from UtrnR+/- mice was quantified by RT-qPCR. f, j, k Expression levels were normalised to 18 S and Tbp and are shown relative to the skeletal muscle sample. Bar graphs show mean + /- SEM.

Fig 4: ERK inhibitors combined with EZH2 inhibitors increased Utrn expression in adult myoblasts.a Schematic of the ERK signalling pathway and the inhibitors used in this study. b Bioluminescence image (left) and quantification of flux (right) of U22 A5 myoblasts after 24 h treatment with EZH2i GSK503 (grey triangles) or ERKi LY32 (red open circles) alone or in combination (black closed circles). LY32 treatment caused a dose-dependent increase in flux which was further amplified by the addition of GSK503 (shaded area). c As in b, for U22 A5 myoblasts treated with EZH2i GSK503 (grey triangles) and ERKi Ravox (red open circles) either alone or combined (black closed circles). A dose dependent increase in flux was observed after Ravox treatment and this was increased upon the addition of GSK503 (shaded area). d RT-qPCR of U22 A5 myoblasts after 24 h treatment with GSK503 and ERK pathway inhibitors (at 10 µM), applied either alone or combined; increased Utrn mRNA upon treatment with ERK pathway inhibitors was amplified by the addition of GSK503. One-way ANOVAs (in order: p = 0.0012, p = 0.0047, p = 0.0036, p = 0.0015) with Sidak’s multiple comparisons test was used to compare between groups. e, f Dmd-deficient U22 A5 myoblast lines 2a6 and 3b2 were generated using CRISPR/Cas9 (details in Fig. S5). UtrnR bioluminescence (blue green) after treatment with GSK503 or LY32 (10 µM) alone or in combination is shown in e and RT-qPCR analysis of Utrn mRNA expression is shown in f. For each clone, comparisons between treatment groups were performed with one-way ANOVAs (e, p = 0.0009 both, f p < 0.0001 and p = 0.0023) with Dunnett’s multiple comparisons test. Expression levels were normalised to 18 S and Tbp and shown relative to the vehicle treated control (d, f). Bars show mean + /- SEM with adjusted p values: *p < 0.05, **p < 0.01, ***p < 0.001) (d–f).

Fig 5: Generation of a bioluminescent screening platform to identify compounds that increase Utrn expression in adult myoblasts.a Schematic showing the generation of UtrnR+/- adult myoblasts. Skeletal muscle was isolated from a heterozygous (UtrnR+/-) female mouse (P22). Muscle tissue was minced and partially digested to encourage outgrowth of myoblasts. Myoblasts were enriched from other cell types by preplating on collagen, immortalised by transfection with SV40 TAg, and clonal lines isolated and used to generate screening platforms. Image created with BioRender.com. b Bioluminescent-based screen for compounds that increase Utrn expression. UtrnR+/- myoblasts (U22 A5) were treated with decreasing concentrations of a panel of epigenetic drugs and kinase inhibitors for 24 h (see Table S1 for full list). At 10 µM, EZH2i (GSK503 and GSK343), BRD9i (GSK602), BRPFi (GSK959), RAFi (LY3009) and ERKi (LY32 and Ravox) (highlighted in yellow) treatment increased Utrn expression as visualised by an increase in flux (p/s). Bars show flux relative to the vehicle only samples. One-way ANOVAs (10 µM p < 0.0001, 1 µM p < 0.0001, 100 nM p = 0.03) with Dunnett’s multiple comparisons test were used to compare vehicle with treated samples. c Utrn mRNA levels measured by RT-qPCR in U22 A5 myoblasts after treatment with drugs showed that 5-Aza and TSA did not increase Utrn, while Utrn was increased by the EZH2 inhibitor GSK503, bromodomain inhibitors (GSK602 and GSK959) and ERK pathway inhibitors (LY3009, LY32, Ravox). One-way ANOVAs (in order: p = 0.07, p = 0.003, p = 0.008, p = 0.013) with Sidak’s multiple comparisons test were used to compare vehicle with treated samples. d–f A myoblast cell line (E22) with conditional deletion of EZH2 was generated from a P22 Ezh2flx/flx male mouse (details are shown in Fig. S4a). d PCR showing removal of the WT allele and accumulation of the deleted allele in E22 myoblasts after treatment with 200 nM 4-hydroxytamoxifen (4-OHT) for 24- 96 h. Primer locations are shown in Fig. S4a. e RT-qPCR showing Ezh2 (upper panel) and Utrn (lower panel) mRNA expression after 72 h 200 nM 4-OHT treatment in E22 myoblasts. Increased Utrn was observed upon Ezh2 depletion. The comparison between vehicle and 4-OHT treatment was performed using a paired t-test. f Western blot of E22 myoblasts after 72 h 200 nM 4-OHT treatment for utrophin, EZH2, H3K27me3, GAPDH and total H3. After 4-OHT treatment, EZH2 and H3K27me3 decreased while utrophin increased. Three replicates are shown, quantification is shown in Fig. S4c. The molecular weight (kDa) of the standards is shown alongside. Full blots are shown in Supplementary data Fig. 2. Expression levels (c, e) were normalised to 18 S and Tbp and are shown relative to the vehicle treated control. Bars represent mean + /- SEM. b, c, e Adjusted p values are shown: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.