Bap1/SMN axis in Dpp4+ skeletal muscle mesenchymal cells regulates the neuromuscular system

Abstract

The survival of motor neuron (SMN) protein is a major component of the pre-mRNA splicing machinery and is required for RNA metabolism. Although SMN has been considered a fundamental gene for the central nervous system, due to its relationship with neuromuscular diseases, such as spinal muscular atrophy, recent studies have also revealed the requirement of SMN in non-neuronal cells in the peripheral regions. Here, we report that the fibro-adipogenic progenitor subpopulation expressing Dpp4 (Dpp4+ FAPs) is required for the neuromuscular system. Furthermore, we also reveal that BRCA1-associated protein-1 (Bap1) is crucial for the stabilization of SMN in FAPs by preventing its ubiquitination-dependent degradation. Inactivation of Bap1 in FAPs decreased SMN levels and accompanied degeneration of the neuromuscular junction, leading to loss of motor neurons and muscle atrophy. Overexpression of the ubiquitination-resistant SMN variant, SMNK186R, in Bap1-null FAPs completely prevented neuromuscular degeneration. In addition, transplantation of Dpp4+ FAPs, but not Dpp4- FAPs, completely rescued neuromuscular defects. Our data reveal the crucial role of Bap1-mediated SMN stabilization in Dpp4+ FAPs for the neuromuscular system and provide the possibility of cell-based therapeutics to treat neuromuscular diseases.

Keywords: Cell Biology; Mouse models; Muscle; Muscle Biology; Neurodegeneration.

Figure 1. Severe muscular atrophy in Bap1^ΔMPC mice.

(A) Gross morphology of tibialis anterior (TA), gastrocnemius (GA), quadriceps (Q), and triceps (Tri) muscles from 19-month-old Bap1^WT and Bap1^ΔMPC mice. (B) Representative IHC staining images of laminin. (C) Morphometric quantifications of CSA of TA and GA myofibers from Bap1^WT and Bap1^ΔMPC mice. (D) Quantification of myofiber numbers in whole TA and GA. (E) Representative IHC staining images of TA muscles of 1-, 3-, and 8-week-old Bap1^WT or Bap1^ΔMPC mice. (F) Morphometric quantifications of CSA of TA muscles. n = 4 animals per group; mean ± SEM; Mann-Whitney U test; *P < 0.05, **P < 0.01. Scale bars: 100 μm (A, B, and E).

Figure 2. Impaired motor functions and NMJs in Bap1^ΔMPC mice.

(A) The score of negative geotaxis test (38). (B and C) hind limb clasping time (85) (B) and immobility time (89) (C) during tail suspension test of 3-week-old Bap1^WT and Bap1^ΔMPC mice. n = 5 animals per group; mean ± SEM; Mann-Whitney U test; *P < 0.05, ***P < 0.001. (D) Representative captures of Bap1^WT and Bap1^ΔMPC mice during tail suspension test. Arrow indicates the hind limb clasping. (E and F) Representative graph data (E) and quantifications of amplitude and latency (F) of CMAP from 1-week-old GA muscles of Bap1^WT and Bap1^ΔMPC mice. (G) Confocal images for BTX and neurofilament immunofluorescence in TA muscles. White arrowheads and arrows indicate swelling of neurofilament and denervation, respectively. (H) Relative mRNA expressions of each AChR subunit in TA muscles. (F and H) n = 4 animals per group; data are mean ± SEM; Mann-Whitney U test; *P < 0.05, **P < 0.01, ***P < 0.001. Scale bars: 2 cm (D) and 20 μm (G).

Figure 3. Loss of motor neurons in Bap1^ΔMPC mice.

(A) Representative electron microscopy images of myelinated axons in sciatic nerve cross sections from 2- (left panels) or 4-week-old (right panels) Bap1^WT and Bap1^ΔMPC mice. (B) The g-ratio of sciatic nerves from 2- or 4-week-old Bap1^WT and Bap1^ΔMPC mice. (C and D) Quantifications of motor neuron (MN) number (C) and representative IHC images of choline acetyltransferase (ChAT) and NeuN in the ventral horn regions of L5 spinal cord from 3- or 8-week-old Bap1^WT and Bap1^ΔMPC mice (D). n = 4 animals for each group (B and C); data are mean ± SEM; Mann-Whitney U test; *P < 0.05, ***P < 0.001. Scale bars: 10 μm (A) and 50 μm (D).

Figure 4. Prevention of neuromuscular degeneration by FAPs.

(A) Representative captures from tail suspension test of Bap1^WT (Cont), PBS-treated (PBS), or FAP-transplanted (FAPs) Bap1^ΔMPC mice. See also Supplemental Video 2. (B and C) Quantifications of hind limb clasping time (B) and immobility time (C) during tail suspension test for 5 minutes. (D) Representative graph data of CMAP. (E and F) Quantifications of amplitude (E) and latency (F) of CMAP measured in GA muscles of Cont, PBS, and FAPs mice. (G) Confocal images of BTX and NF immunofluorescence (top) and IHC images of TA muscles (bottom). (H) Percentages of the innervated NMJs. (I and J) Representative IHC images of ChAT and neurofilament (NF) in ventral horn regions of L5 spinal cord (I) and quantifications of MN number (J). (K) Quantification of CSA. n = 5 animals for each group (B, C, E, F, H, J, and K) (n = 6 animals in J); data are mean ± SEM; Tukey’s pairwise comparison test after 1-way ANOVA; *P < 0.05, **P < 0.01, ***P < 0.001. Scale bars: 2 cm (A), 20 μm (G, top), 100 μm (G, bottom), and 50 μm (I).

Figure 5. Maintenance of the neuromuscular system by FAPs.

(A) Experimental scheme of FAPs’ transplantation. FAPs from hind limb muscles of 1.5-week-old Rosa-CreER Bap1^fl/fl mice were transplanted into TA and GA muscles of 1.5-week-old Bap1^ΔMPC mice. Eight weeks after transplantation, the transplanted mice were orally administered with tamoxifen (Tmx) for 3 consecutive days. (B and C) Tail suspension test was performed at the indicated weeks after Tmx administration. Representative captures of tail suspension tests (B). Arrow indicates the hind limb clasping. See also Supplemental Video 3. Quantification of immobility time during tail suspension test for 5 minutes (C). n = 5 animals for each group; data are mean ± SEM; Tukey’s pairwise comparison test after 1-way ANOVA; *P < 0.05, **P < 0.01. (D–G) Histological analyses were performed at the indicated weeks after Tmx administration. Representative confocal images of NMJs (D) and quantification of innervated NMJs (E). Representative IHC images of ChAT (F) and quantifications of MNs (G) in L5 spinal cords. (H) Representative IHC staining images of TA muscles. (I) Quantification of CSA in TA muscles. (E, G, and I) n = 5 animals for each group; data are mean ± SEM; unpaired t test; **P < 0.01, ***P < 0.001. Scale bars: 2 cm (B), 25 μm (D), 50 μm (F), and 100 μm (H).

Figure 6. Identification of the neuroprotective Dpp4⁺ subpopulation within FAPs.

(A) Uniform manifold approximation and projection (UMAP) plot showing distinct subpopulations (Dpp4⁺ or Cxcl14⁺) within Bap1^WT versus Bap1^ΔMPC FAPs. (B) Heatmap of top 100 differentially expressed genes in Bap1^WT versus Bap1^ΔMPC FAPs. (C) Heatmap showing expression of marker genes used to distinguish the 2 subpopulations in each cluster. (D) GO analysis using statistical overrepresentation test on DEGs selected from comparing Bap1^WT versus Bap1^ΔMPC FAPs in each subpopulation. (E) Representative FACS plots of Dpp4⁺ versus Dpp4^– FAPs isolated from 4-week-old tdTomato reporter mice. (F) Quantification of immobility time during tail suspension test. Bap1^ΔMPC mice transplanted with PBS, Dpp4⁺ FAPs, or Dpp4^– FAPs were compared with Bap1^WT mice. n = 5 animals for each group; data are mean ± SEM; Tukey’s pairwise comparison test after 1-way ANOVA; **P < 0.01. (G and H) Representative IHC images of ChAT (scale bars: 50 μm) (G) and quantifications of MN number stained with ChAT on the ventral horn region of L5 spinal cord in Dpp4⁺ FAP– or Dpp4^– FAP–transplanted Bap1^ΔMPC mice (H). n = 5 animals for each group. Mean ± SEM; unpaired t test; ***P < 0.001.

Figure 7. Deubiquitination of SMN by Bap1.

(A) Relative expression of SMN in FAPs from hind limb muscles of 1.5-week-old Bap1^WT and Bap1^ΔMPC mice. Gapdh and b-actin were used for normalization in quantitative PCR analysis. n = 4 animals for each group; Mann-Whitney U test; data are mean ± SEM. (B) Immunoblot of Bap1 and SMN in FAPs from 4-week-old Bap1^WT and Bap1^ΔMPC mice. (C) Immunoblot of Bap1 and SMN after immunoprecipitation (IP) analysis. IP was performed with pre-immune IgG, anti-Bap1, or anti-SMN antibodies. Each sample was blotted with each indicated antibody. (D) Ubiquitination assay of SMN following BAP1 overexpression. Flag-tagged BAP1 was expressed in HEK293T cells with HA-Ubiquitin (HA-Ub) and HisMax-tagged SMN (His-SMN). Cell lysates were subjected to pull-down assays with NTA resins followed by immunoblot. (E) Ubiquitination assay of SMN following BAP1 or BAP1^C91S overexpression. HEK293T cells expressing Flag-tagged BAP1 or its C91S mutant were subjected to IP with anti-SMN antibody. After IP, each sample was blotted with each indicated antibody. (F) Ubiquitination assay of SMN and its lysine-substitution variants. HisMax-tagged variants (K41R, K51R, or K186R) were expressed in HEK293T cells with or without HA-Ub. Cell lysates were subjected to pull-down assays with NTA resins followed by immunoblot with each indicated antibody. (G) Immunoblot of SMN in Bap1^WT or Bap1^ΔMPC FAPs transduced with lentiviral vectors containing His-tagged SMN^WT or SMN^K186R. Samples were blotted with each indicated antibody. (H) Immunoblot analysis of SMN protein half-life. Bap1^ΔMPC FAPs transduced with lentiviral vectors containing His-tagged SMN or its K186R mutant were subjected to immunoblot with indicated antibodies. Cells were incubated with cycloheximide (CHX) for the indicated times. (I) A schematic model for deubiquitination of SMN1 by BAP1. (D, F, and G) For MG132 treatment, cells were incubated with 10 μM MG132 for 4 hours before the preparation of cell lysates.

Figure 8. Rescue of Bap1^ΔMPC mice by SMN^K186R-expressing FAPs.

(A) Representative captures of hind limb clasping during tail suspension test. See also Supplemental Video 5. (B) Quantification of immobility time during tail suspension test for 5 minutes. (C) Representative graph data of CMAP. (D and E) Quantification of amplitude (D) and latency (E) of CMAP on GA muscles. (F) Representative confocal images for BTX and NF (F, top) and IHC staining (F, bottom) of TA muscles. (G) Quantification of innervated NMJs. (H) Representative images of ChAT and NeuN in the ventral horn of L5 spinal cord. (I) Quantification of the number of MNs in L5 spinal cord. (J) Quantification of CSA in TA muscles. (B, D, E, G, I, and J) n = 5 animals for each group; data are mean ± SEM; Tukey’s pairwise comparison test after 1-way ANOVA; *P < 0.05, **P < 0.01, ***P < 0.001. (A–J) tail suspension tests, CMAP measure, and histological analyses were performed at 8 or 10 weeks after transplantation. Scale bars: 2 cm (A), 25 μm (F, top), 100 μm (F, bottom), 50 μm (H).