Muscle Yap Is a Regulator of Neuromuscular Junction Formation and Regeneration

Abstract

Yes-associated protein (Yap) is a major effector of the Hippo pathway that regulates cell proliferation and differentiation during development and restricts tissue growth in adult animals. However, its role in synapse formation remains poorly understood. In this study, we characterized Yap's role in the formation of the neuromuscular junction (NMJ). In HSA-Yap^-/- mice where Yap was mutated specifically in muscle cells, AChR clusters were smaller and were distributed in a broader region in the middle of muscle fibers, suggesting that muscle Yap is necessary for the size and location of AChR clusters. In addition, HSA-Yap^-/- mice also exhibited remarkable presynaptic deficits. Many AChR clusters were not or less covered by nerve terminals; miniature endplate potential frequency was reduced, which was associated with an increase in paired-pulse facilitation, indicating structural and functional defects. In addition, muscle Yap mutation prevented reinnervation of denervated muscle fibers. Together, these observations indicate a role of muscle Yap in NMJ formation and regeneration. We found that β-catenin was reduced in the cytoplasm and nucleus of mutant muscles, suggesting compromised β-catenin signaling. Both NMJ formation and regeneration deficits of HSA-Yap^-/- mice were ameliorated by inhibiting β-catenin degradation, further corroborating a role of β-catenin or Wnt-dependent signaling downstream of Yap to regulate NMJ formation and regeneration.SIGNIFICANCE STATEMENT This paper explored the role of Yes-associated protein (Yap) in neuromuscular junction (NMJ) formation and regeneration. Yap is a major effector of the Hippo pathway that regulates cell proliferation and differentiation during development and restricts tissue growth in adult animals. However, its role in synapse formation remains poorly understood. We provide evidence that muscle Yap mutation impairs both postsynaptic and presynaptic differentiation and function and inhibits NMJ regeneration after nerve injury, indicating a role of muscle Yap in these events. Further studies suggest compromised β-catenin signaling as a potential mechanism. Both NMJ formation and regeneration deficits of HSA-Yap^-/- mice were ameliorated by inhibiting β-catenin degradation, corroborating a role of β-catenin or Wnt-dependent signaling downstream of Yap to regulate NMJ formation and regeneration.

Keywords: YAP; neuromuscular junction; regeneration; β-catenin.

Figure 1.

Decreased muscle strength in HSA-Yap^−/− mice. A, Schematic diagrams of WT, floxed, and deleted alleles of the mouse Yap gene. P1, P2, and P3 were genotyping primers. E1, Exon 1; E2, exon 2. B, Genotyping results of HSA-Yap^+/+; Yap^f/f; Yap^f/+; HSA-Yap^−/− mice. Tail and muscle DNA was isolated and subjected to PCR analysis. Arrows indicate bands expected for different genotypes. Molecular weight markers (in bp) are indicated on the left. C, Muscle-specific reduction of Yap. Homogenates of different tissues were subjected to Western blot with anti-Yap and GAPDH antibodies. D, Decreased grip strength of HSA-Yap^−/− mice. ***p = 0.00061 (unpaired t test). N = 20 mice per group. E, Reduced time in wire hanging test of HSA-Yap^−/− mice. *p = 0.012 (unpaired t test). N = 23 mice per group. F, Increased falling from Rotarod of HSA-Yap^−/− mice. *p < 0.05 (two-way ANOVA). ***p < 0.001 (two-way ANOVA). N = 15 mice per group. G, Comparable levels of myogenic factor mRNA between control and mutant groups. N = 3 or 4 mice per group. H, Representative images of gastrocnemius cross sections. Scale bar, 20 μm. I, Comparable muscle types between control and mutant mice. N = 5 mice per group. J, No difference in cross section area of muscle fibers. N = 6 mice per group. K, Similar nuclear distribution on TA and soleus muscle. Red represents laminin. Blue represents DAPI. Scale bar, 20 μm. L, Quantification of central nuclei percentage of K. N = 5 mice per group.

Figure 2.

Reduced twitch and tetanic force by nerve stimulation in HSA-Yap^−/− mice. A, Scheme of in vivo muscle twitch and tetanic force measurement by muscle stimulation. B, Comparable single twitch force between control and mutant mice by muscle stimulation. N = 4 mice per group. C, Representative tetanic curves at stimulation frequency 50, 100, and 125 Hz by muscle stimulation. D, Comparable tetanic force between control and mutant mice by muscle stimulation. N = 4 mice per group. E, Scheme of in vivo muscle twitch and tetanic force measurement by nerve stimulation. F, Reduced single twitch force in HSA-Yap^−/− mice by nerve stimulation. *p = 0.025 (unpaired t test). N = 4 mice per group. G, Representative tetanic curves at stimulation frequency 50, 100, and 125 Hz by sciatic nerve stimulation. H, Reduced tetanic force in HSA-Yap^−/− mice by nerve stimulation. *p < 0.05 (unpaired t test). N = 4 mice per group.

Figure 3.

Postsynaptic and presynaptic deficits in HSA-Yap^−/− mice. A, Representative images of E15.5 and P0 Yap^f/f and HSA-Yap^−/− diaphragm left ventral region. Muscles were stained with CF568 α-BTX (red) and anti-NF/Syn antibodies (visualized by AlexaFluor-488 goat anti-rabbit IgG, green). Arrow indicates primary branch. Arrowhead indicates defasciculated secondary branch. Scale bar, 100 μm. B, Enlarged images of AChR clusters. Scale bar, 10 μm. C, Increased endplate band width in HSA-Yap^−/− mice. *p = 0.014 for E15.5 (unpaired t test). **p = 0.0018 for P0 (unpaired t test). N = 4 or 5 mice per group. D, Comparable AChR number between Yap^f/f and HSA-Yap^−/− mice. N = 4 mice per group. E, AChR cluster intensity was comparable between Yap^f/f and HSA-Yap^−/− mice. N = 4 or 5 mice per group. F, Reduced AChR area in HSA-Yap^−/− mice. *p = 0.011 for P0 (unpaired t test). N = 4 mice per group. G, Reduced secondary branch number in HSA-Yap^−/− mice. **p = 0.0085 for E15.5 (unpaired t test). **p = 0.0014 for P0 (unpaired t test). N = 4 or 5 mice per group. H, Longer secondary branch in Yap mutant mice. *p = 0.021 for E15.5 (unpaired t test). **p = 0.0037 for P0 (unpaired t test). N = 4 or 5 mice per group. I, Reduced nerve coverage in HSA-Yap^−/− NMJs. ***p = 8.2E-5 for E15.5 (unpaired t test). **p = 0.0024 for P0 (unpaired t test). N = 4 or 5 mice per group.

Figure 4.

CMAP reduction in HSA-Yap^−/− mice. A, Representative CMAP traces of Yap^f/f and HSA-Yap^−/− mice in response to first, second, and 10th stimuli at 40 Hz. B, Ten CMAP traces shown in a stacked succession for comparison. C, D, Reduced CMAP amplitudes. CMAP amplitude ratio of the 10th to the first traces at different stimulation frequency (C) and reduced CMAP amplitude at 40 Hz (D). *p < 0.05 (two-way ANOVA). **p < 0.01 (two-way ANOVA). ***p < 0.001 (two-way ANOVA). N = 6 mice per group.

Figure 5.

Impaired neuromuscular transmission in HSA-Yap^−/− mice. A, Comparable resting membrane potentials between Yap^f/f and HSA-Yap^−/− mice at P3 and P30. N = 6 mice per group. B, Representative mEPP traces of P3 (top) and P30 (bottom) of Yap^f/f and HSA-Yap^−/− mice. Underlined regions in the left were enlarged in the right. C, D, Normal mEPP amplitudes (C) but reduced mEPP frequency (D) in HSA-Yap^−/− mice. **p = 0.0015 for P3 mEPP frequency (unpaired t test). ***p = 9.1E-6 for P30 mEPP frequency (unpaired t test). N = 6 mice per group. E, Reduced EPP amplitudes in HSA-Yap^−/− mice. *p = 0.016 (unpaired t test). N = 6 mice per group. F, Representative paired-pulse traces at 10 ms of stimulation interval. G, Increased paired-pulse facilitation in HSA-Yap^−/− mice. ***p < 0.001 (two-way ANOVA). N = 4 mice per group.

Figure 6.

Reduced nerve terminal proteins in HSA-Yap^−/− mice. A, B, No difference of agrin (A) and S100β (B) staining in gastrocnemius between Yap^f/f and HSA-Yap^−/− mice. Muscles were stained whole-mount with respective antibodies. Scale bar, 10 μm. C, D, α-BTX area covered by agrin (C) and S100β (D). N = 5 or 6 mice per group. E, F, Quantification of pretzel-like structures at P30. Comparable AChR fragment number (E) and perforation (F) between Yap^f/f and HSA-Yap^−/− mice. N = 5 mice per group. G, Reduced SV2 and synapsin protein in synaptic region of gastrocnemius. H, Quantification of data in G. *p = 0.019 for P3 SV2 (unpaired t test). **p = 0.0059 for P30 SV2 (unpaired t test). *p = 0.017 for P30 synapsin (unpaired t test). N = 3 mice per group. I, J, Reduced SV2 (I) and synapsin (J) staining at mutant NMJ. Gastrocnemius was stained whole-mount with respective antibodies. Scale bar, 10 μm. K, Reduced α-BTX area covered by SV2 and synapsin. **p = 0.0013 for P3 SV2 (unpaired t test). **p = 0.0014 for P3 synapsin (unpaired t test). **p = 0.0021 for P30 SV2 coverage (unpaired t test). ***p = 2.4E-7 for P30 synapsin coverage (unpaired t test). N = 5 or 6 mice per group.

Figure 7.

Compromised NMJ regeneration in HSA-Yap^−/− mice. A, Schematic diagram of tibial nerve transplant surgery. Both medial and lateral gastrocnemius muscles were denervated by removing a 5 mm segment immediately before the entrance to respective muscles. The tibial nerve was severed before it branches into lateral and medial plantar nerves, and its proximal end was transplanted and sutured ∼2 mm above the midline synaptic region of the lateral gastrocnemius. As control, the denervated medial gastrocnemius was not transplanted with a nerve (see Materials and Methods). B, Representative images of NMJ regeneration in lateral (left panel), but not medial (right panels) gastrocnemius, 4 weeks after nerve transplant surgery. White arrowheads indicate fully reinnervated AChR clusters. Blue arrowheads indicate partially reinnervated or denervated AChR clusters. Scale bar, 100 μm. C, Enlarged images of newly generated NMJs. Scale bar, 10 μm. D, Decreased fully reinnervated and increased partially reinnervated AChR clusters in HSA-Yap^−/− mice. Den, Denervated AChR clusters; Partial Inn, partially reinnervated AChR clusters; Full Inn, fully reinnervated AChR clusters. **p < 0.01 (two-way ANOVA). N = 4 mice per group. E, Representative images of NMJ regeneration in gastrocnemius 14 and 21 d after nerve crush. White arrowheads indicate fully reinnervated AChR clusters. Blue arrowheads indicate partially reinnervated or denervated AChR clusters. Scale bar, 200 μm. F, Reduced fully reinnervated and increased denervated AChR clusters in HSA-Yap^−/− mice after nerve crush. ***p < 0.001 for denervated and partially reinnervated AChR clusters 14 d after crush (two-way ANOVA). **p < 0.01 for fully innervated AChR clusters 21 d after crush (two-way ANOVA). N = 4 mice per group.

Figure 8.

Reduced β-catenin and Slit2 in HSA-Yap^−/− mice. A, β-Catenin was reduced in HSA-Yap^−/− gastrocnemius. B–E, Quantification of β-catenin (B), δ-catenin (C), MuSK (D), and AChR β-subunit (E) data in A. B, *p = 0.016 for P0 (unpaired t test). **p = 0.0011 for P10 (unpaired t test). ***p = 0.00077 for P30 (unpaired t test). N = 3 mice per group. F, β-Catenin was reduced at both nuclear and cytoplasmic fraction in P30 HSA-Yap^−/− gastrocnemius. G, Quantification of data in F. ***p = 0.00052 for cytoplasmic fraction (unpaired t test). **p = 0.0055 for nuclear fraction (unpaired t test). N = 3 mice per group. H, mRNA levels of factors implicated in NMJ formation. *p = 0.029 for Slit2 (unpaired t test). *p = 0.031 for Gdnf (unpaired t test). **p = 0.0067 for Gdf5 (unpaired t test). **p = 0.0045 for Fgf10 (unpaired t test). N = 3 or 4 mice per group.

Figure 9.

Partial rescue of NMJ formation and regeneration deficits by LiCl treatment. A, Representative images of diaphragm left ventral region of different mice. Muscles were stained with CF568 α-BTX (red) and anti-NF/Syn antibodies (visualized by AlexaFluor-488 goat anti-rabbit IgG, green). Arrow indicates primary branch. Arrowhead indicates defasciculated secondary branch. Scale bar, 100 μm. B, Enlarged images of AChR clusters and nerve terminals. Scale bar, 10 μm. C, D, No effect of LiCl on reduced AChR area (C) or increased endplate band width (D). E, F, No effect on secondary branch number was observed for LiCl, but secondary branch length was reduced in LiCl-treated HSA-Yap^−/− mice compared with saline-treated mutant mice. ***p = 3.9E-5 for group Yap^f/f versus HSA-Yap^−/−-saline (one-way ANOVA). *p = 0.028 for group HSA-Yap^−/−-saline versus HSA-Yap^−/−-LiCl (one-way ANOVA). N = 6 mice per group. G, Nerve coverage was increased in LiCl-treated mutant mice compared with saline-treated mutant. ***p = 0.00019 for group Yap^f/f versus HSA-Yap^−/−-saline (one-way ANOVA). **p = 0.0023 for group HSA-Yap^−/−-saline versus HSA-Yap^−/−-LiCl (one-way ANOVA). N = 6 mice per group. H–J, mEPP frequency in HSA-Yap^−/− mice was increased after LiCl treatment. H, Representative mEPP traces. Underlined regions in the left were enlarged in the right. I, Reduced mEPP frequency was rescued by LiCl treatment. ***p = 0.00012 for group Yap^f/f versus HSA-Yap^−/−-saline (one-way ANOVA). **p = 0.0058 for group HSA-Yap^−/−-saline versus HSA-Yap^−/−-LiCl (one-way ANOVA). N = 6 mice per group. J, mEPP amplitude was comparable among all groups. K, L, More fully reinnervated AChR clusters in LiCl-treated HSA-Yap^−/− mice compared with saline-treated mutant mice. K, Representative images of regenerated NMJs. White arrowhead indicates fully reinnervated AChR clusters. Blue arrowhead indicates partially reinnervated or denervated AChR clusters. Scale bar, 100 μm. L, More fully innervated AChR clusters in LiCl-treated mutant mice. *p < 0.05 (two-way ANOVA). **p < 0.01 (two-way ANOVA). ***p < 0.001 (two-way ANOVA). N = 4 mice per group.