Dystrophic fibres in lama2^−/− larvae undergo extensive remodelling prior to undergoing cell death. a Still frames of continuous time-lapse analysis (Supplementary Movie 1) of mosaic labelled actc1b:GFP dystrophic fibres in lama2^−/− larvae reveals extensive remodelling and reattachment (blue arrows) to the matrix-rich regions of the vertical myosepta (red dashed lines). Time (t) is indicated in minutes. b Tracing of single dystrophic fibres in lama2^−/− and dmd^−/− larvae reveals lama2^−/− dystrophic fibres are long lived compared to those evident dmd^−/− animals. c, d Fibre remodelling is evident in dystrophic fibres in lama2^−/− larvae mosaically injected with an act1b:H2az2a-mCherry-IRES-EGFP-CaaX construct, which marks differentiated muscle nuclei in red and plasma membrane in green. This analysis identified extensive fibre branching (arrows, d) of an individual dystrophic fibre that was not evident in wildtype siblings similarly injected (c). e, f Fibres transgenically marked mosaically with the alpha-actinin-mKate2 fusion protein expressed shows the formation of new pre-myofibrils in the dystrophic context (f), compared to Z-line associated in the wildtype (e). g–i Dystrophic fibres in lama2^−/− animals undergo a delayed cell death. An increase in cell death is only seen at 5 dpf within the myotome of lama2^−/− larvae, several days after the onset of dystrophic pathology. g, h″ TUNEL staining (Green) in wildtype sibling (g) and lama2^−/− larvae (h–h″) transgenic for Tg(actc1b:H2az2a-mCherry), which marks muscle nuclei in red, reveals the increase in cell death in muscle nuclei in lama2^−/−. h′ and h″ are high magnification views of the region marked in h. i Quantitation of TUNEL-positive muscle nuclei in the myotomes of lama2^−/− animals. Significance is tested in a one-way ANOVA, Tukey’s test; ***p < 0.001. Results are expressed as a mean ± SD. j Secreted annexin5-mKate2 (red, arrows), which marks cells undergoing cell death, is specific for regions of fibre loss in 5 dpf larvae. Blue marks intact muscle fibres in Tg(actc1b:EBFP2) larvae. Scale bars, 20 μm

Long-lived dystrophic lama2^−/− fibres maintain fibre polarity, upregulate Dystrophin and undergo hyper-fusion. a–c Dystrophin expression (yellow) in the line GT(dmd-citrine), is mislocalised in lama2^−/− mutants (b) compared to WT (a) in confocal projections. Single confocal section (c) reveals junctional polarity of Dystrophin is maintained in lama2^−/− fibres (indicated by arrowhead). Muscle is marked in blue by BFP from Tg(actc1b:EBFP2). a, b Maximum projections from Supplementary Movies 2 and 3. dlama2^−/− dystrophic fibres undergo fusion during remodelling. Injection of actc1b:H2az2a-EosFPtd-IRES-EGFP-CaaX into lama2^−/− embryos allows photoconversion of muscle nuclei expressing H2az2a-EosFPtd from green to red at 3 dpf (arrrowheads). Fibres are tracked and addition of new nuclei (green, arrows) determined. This reveals fusion which precedes reattachment of the fibre to the opposite vertical myoseptum. Time is expressed as days post-photocoversion. e–k Hyper-fusion in lama2^−/− dystrophic fibres. e, f Maximum projections of WT sibling (e) and lama2^−/− (f) larvae transgenic for Tg(actc1b:H2az2a-mCherry) and Tg(actc1b:EBFP2), which mark muscle nuclei (red) and muscle (blue), respectively. Insets reveal the regular arrays of nuclei in WT are perturbed and clumped in lama2^−/− fibres. g Confocal section of dystrophic fibres within a lama2^−/− mutant transgenic for Tg(actc1b:H2z2a-mCherry) and Tg(actc1b:EGFP-CaaX). Hyper-fused fibres are evident, with the centrally located fibre exhibiting clumped nuclei, a situation never seen in wildtypes. h–k Activation of the stem cell compartment as a source of muscle nuclei. WT (h, j) or lama2^−/− (i, k) larvae transgenic for Tg(cmet:KalTA4; UAS-nlsGFP) and Tg(actc1b:EBFP2). Green marks the cmet-positive satellite cell compartment (54) and blue differentiated fibres, revealing the GFP-positive satellite cell-derived nuclei, specifically in lama2^−/− dystrophic fibres (arrowheads i, k inset). Arrows mark the lateral line and brackets indicate the cell body nuclei of the dorsal motor neurons, which also express cmet at high level (h, i). All panels of 3 dpf larvae unless otherwise denoted. l–n Quantitation of the hyper-fusion phenotype in lama2^−/− mutant larvae. Phalloidin (red) marks muscle fibres and DAPI (green) marks nuclei in wildtype sibling (l) and mutant larvae (m). Quantitation of the myonuclear domain size, which measures that relative amount muscle cell cytoplasm to muscle nuclei within individual fibres (n). More nuclei per unit area is represented by a smaller myonuclear domain size. Significance is tested in a one-way ANOVA, Tukey’s test; ****p < 0.0001. Scale bars, 20 μm

Transgenic delivery of laminin alleviates dystrophic pathology in lama2^−/− animals. a, b Germline transgenic larvae expressing Tg(actc1b:Lama2-mCherry) results in correct Lama2-mCherry (red) deposition at the myosepta in WT (a) and lama2^−/− deficient (b) larvae and rescues the dystrophic pathology in lama2^−/− animals. c Germline transgenic lama2^−/− larvae carrying the Tg(hsp70l:Lama2-mCherry) cassette are rescued upon heat-shock-induced expression of the Lama2-mCherry fusion protein (red). Blue marks differentiated muscle fibres in the Tg(actc1b:EBFP2). The focal “dot-like” Lama2-mCherry localisation results from the non-tissue-specific heat shock induction strategy, which also induces expression in non-myotomal cells. In this image, non-target expression is most prominent in cells of the epidermis. d–f Mosaic overexpression of Lama2 can rescue pathology in lama2^−/− larvae. Injection of actc1b:Lama2-mCherry-IRES-GFP-CaaX into lama2^−/− (d) reveals that even a low level of mosaic transgenesis (secreting transgenic cells labelled by localisation of GFP to the membranes) results in correct localisation to the myosepta, and a rescue of pathology (g). By contrast a mutant, non-polymerising (NP) form of Lama2, when expressed in the same manner (actc1b:Lama2-NPmCherry-IRES-GFP-CaaX), fails to accumulate at the myosepta in WT larvae (e) and does not rescue lama2^−/− mutants (f). g Injection of actc1b:Lama2-mCherry-IRES-GFP-CaaX into dmd^−/− larvae does not rescue muscle integrity. By contrast similar experiments using lama2^−/− larvae results in an amelioration of muscle pathology. Results are expressed as a mean ± SD, each group n = 12 per data point. Significance is tested in a two-way ANOVA between groups at each time point; *p < 0.05, ****p < 0.0001. Scale bars, 20 μm

Laminin can rescue pathology in lama2^−/− larvae. a Heat shocked transgenic lama2^−/− larvae carrying the Tg(hsp70l:Lama2-mCherry) cassette are were measured for active force at 6 dpf using a force transducer. The optimal length for active contraction was first determined, then muscles were stimulated at this length at supra-maximal voltage for maximal force production. Providing heatshock at 3 but not at 5 dpf resulted in an increased active force within mutants. b, c Two distinct methods of exogenous delivery of fluorescently labelled Lam111, systemic (b) and direct intramuscular injection (c) lead to the accumulation of laminin protein preferentially at areas of fibre loss (arrows), as detected in Tg(actc1b:EBFP2) as reduced blue fluorescence. d, e Injection into the transgenic line Tg(smyhc1:GFP) which provides single-fibre resolution of the superficial slow muscle palisade, reveals both focal (arrowheads) and large fibre-associated regions (arrows). b, c Single confocal plane images; d, e maximum projection images of myotome. Larvae at 5 dpf. f Injection of Lam111 at 3 dpf ameliorates the rate of muscle loss in lama2^−/− larvae. Results are expressed as a mean ± SD, each group n = 12 per data point. Significance is tested in a two-way ANOVA between groups at each time point; *p < 0.05, ****p < 0.0001. Scale bars, 20 μm

lama2^−/− mutant larvae possess defects in muscle stem and progenitor cell formation that are rescued by Laminin protein injection. a–c The transgenic line TgBAC(pax3a:GFP) marks muscle stem and progenitor cells in zebrafish larvae. lama2^−/− mutants (b) possess reduced numbers of pax3a-GFP-positive cells compared to WT (a), quantitated in c (Results expressed as mean ± SD. For each group, n ≥ 11. Statistics: t-test, two-tailed; ****p < 0.0001). d, e Intramuscular injection of fluorescently labelled Laminin protein (red) into lama2^−/− mutants (e) rescues pax3a-GFP-positive cell numbers, and increases them significantly over WT levels (d), quantitated in f (Results expressed as mean ± SD. For each group, n ≥ 7. Statistics: one-way ANOVA, Tukey’s test; *p < 0.05, **p < 0.01). Scale bars, 20 μm

Laminin stimulation of stem cell proliferation requires downstream integrin signalling. a, b Representative images of phosphorylated-FAK expression (green) within skeletal muscle of 6 dpf lama2^−/− or sibling zebrafish. Scale bar, 20 μm. c Quantification of expression levels of p-FAK at the myotendious junctions of 6 dpf lama2^−/− or sibling zebrafish. Results are expressed as mean ± SD. Statistics: t‐test, two tailed; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. d–i Representative images of p-FAK expression (green) in 6 dpf lama2^−/− zebrafish treated with laminin protein (red) or PBS. Note the colocalization of both these proteins at the myotendious junctions. Scale bar, 20 μm. j Quantification of expression levels of p-FAK at the myotendious junctions of 6 dpf lama2^−/− zebrafish that were either treated with exogenous laminin, PBS (orange) or untreated. Results are expressed as a mean ± SD. Statistics: one-way ANOVA, Tukey’s test; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. k, l Representative images of Laminin treatment in the ilk^−/− mutant. White arrow indicates where intramuscular injection took place. Muscle stem cells marked by TgBAC(pax3a:GFP). Scale bar, 20 μm. m Quantitative analysis shows there is no significant effect on the ilk^−/− mutant when laminin treated; ilk^−/− mutant zebrafish were either treated with exogenous laminin or PBS. Results are expressed as a mean ± SD. For each group, n ≥ 6. Statistics: one-way ANOVA, Tukey’s test; *p < 0.05, **p < 0.01. Scale bars, 20 μm