SMN Deficiency Induces an Early Non-Atrophic Myopathy with Alterations in the Contractile and Excitatory Coupling Machinery of Skeletal Myofibers in the SMN∆7 Mouse Model of Spinal Muscular Atrophy

Abstract

Spinal muscular atrophy (SMA) is caused by a deficiency of the ubiquitously expressed survival motor neuron (SMN) protein. The main pathological hallmark of SMA is the degeneration of lower motor neurons (MNs) with subsequent denervation and atrophy of skeletal muscle. However, increasing evidence indicates that low SMN levels not only are detrimental to the central nervous system (CNS) but also directly affect other peripheral tissues and organs, including skeletal muscle. To better understand the potential primary impact of SMN deficiency in muscle, we explored the cellular, ultrastructural, and molecular basis of SMA myopathy in the SMNΔ7 mouse model of severe SMA at an early postnatal period (P0-7) prior to muscle denervation and MN loss (preneurodegenerative [PND] stage). This period contrasts with the neurodegenerative (ND) stage (P8-14), in which MN loss and muscle atrophy occur. At the PND stage, we found that SMN∆7 mice displayed early signs of motor dysfunction with overt myofiber alterations in the absence of atrophy. We provide essential new ultrastructural data on focal and segmental lesions in the myofibrillar contractile apparatus. These lesions were observed in association with specific myonuclear domains and included abnormal accumulations of actin-thin myofilaments, sarcomere disruption, and the formation of minisarcomeres. The sarcoplasmic reticulum and triads also exhibited ultrastructural alterations, suggesting decoupling during the excitation-contraction process. Finally, changes in intermyofibrillar mitochondrial organization and dynamics, indicative of mitochondrial biogenesis overactivation, were also found. Overall, our results demonstrated that SMN deficiency induces early and MN loss-independent alterations in myofibers that essentially contribute to SMA myopathy. This strongly supports the growing body of evidence indicating the existence of intrinsic alterations in the skeletal muscle in SMA and further reinforces the relevance of this peripheral tissue as a key therapeutic target for the disease.

Keywords: SMA; SMN∆7 mice; actin filaments; mitochondria; sarcomere; skeletal muscle; triads.

Figure 1

(A) Quantitative analysis of the righting motor reflex acquisition by WT (n = 6) and SMN∆7 (n = 6) mice at the indicated postnatal ages. p values from WT and SMN∆7 data comparison was 0.2516, 3.1 × 10⁻¹⁰, 1.7 × 10⁻¹⁰, and 5.5 × 10⁻¹¹ at P0, P5, P10, and P14, respectively. (B) Quantitative analysis of the mean myofiber diameters from WT (n = 5) and SMNΔ7 (n = 5) TA muscle at the indicated postnatal ages. At least 500 measurements per experimental group were performed using the Image J software (v.13.0.6. National Institutes of Health, USA) on transversal cryosections stained with FITC-phalloidin. p values from WT and SMN∆7 data comparison was 0.0618, 0.0647, 1.5× 10⁻⁵, and 8.9× 10⁻⁶ at P0, P5, P10, and P14, respectively. (C–J) Representative confocal microscopy images of transversal cryosections of TA muscle, stained with FITC-phalloidin, used for the quantification of the mean myofiber diameter shown in panel B. Images from WT (C–F) and SMNΔ7 (G–J) mice at P0 (C,G), P5 (D,H), P10 (E,I), and P14 (F,J). In all graphs (A,B), values are shown as mean ± SD, and unpaired Student’s t test was used for statistical analysis; ***: p < 0.0005. Scale bar: 10 µm (C–J).

Figure 2

(A–D) qRT-PCR determination of the normalized expression levels of IL-15 (A), MuRF1 (B), Atrogin-1 (C), and Dcn (D) mRNAs in TA extracts from P5 SMN∆7 mice (n = 5) relativized to age-matched WT animals (n = 3). p values from WT and SMN∆7 data comparison was 0.5588 for IL-15, 0.4011 for MuRF1, 0.4919 for Atrogin-1, and 0.0005 for Dcn. (E) Electron micrograph illustrating a muscle satellite cell (SC) in telophase closely attached to a myofiber in a transversal section. (F–I) qRT-PCR determination of normalized mRNA expression levels of Pax7 (F), MyoD (G), Myog (H), and Mrf4 (I) in RNA extracts of TA muscle samples from SMN∆7 mice (n = 5) and age-matched WT animals (n = 3) at P5. Note in SMN∆7 samples the approximately five-fold increase of Pax7 mRNA expression at the PND stage. p values from WT and SMN∆7 data comparison were 0.00296 for Pax7, 0.3924 for MyoD, 0.7655 for Myog, and 0.5049 for Mrf4. In all graphs (A–D and F–I), values are shown as mean ± SD, and unpaired Student’s t test was used for statistical analysis; **: p < 0.005, ***: p < 0.0005. Scale bar: 2 µm (E).

Figure 3

(A–E) Representative confocal images of longitudinal cryosections of the TA muscle stained with FITC-phalloidin from WT at P5 (A) and SMN∆7 mice at P0 (B), P5 (C,D), and P14 (E). Note in SMN∆7 mice images the presence of myofiber regions with disruption of cross striation and aberrant accumulations of F-actin showing bright FITC-phalloidin fluorescent signal. (F–H) Double staining to label actin filaments (Phalloidin, green channel) and nuclei (PI, red channel) in WT (F) and SMN∆7 (G,H) myofibers at P5. Note the typical cross striation and peripheral positioning of myonuclei in WT myofibers (F). (G,H) Asteriks indicatethe presence of bright F-actin accumulation, some of them closely associated with central or peripheral myonuclei in SMN∆7 myofibers. Scale bars: 20 µm (A–G) and 10 µm (H).

Figure 4

(A–H) Representative electron micrographs of longitudinal sections of TA myofibers from WT (A,F) and SMN∆7 (B–E,G,H) mice. (A,F) Normal fine structure and organization of sarcomeres in WT myofibers at P5 (A) and P14 (F). (B–E,G,H) Ultrastructural alterations of the contractile machinery with misalignment of Z discs (labeled as “Z”) and presence of local sarcoplasmic areas of variable size with disruption of sarcomere architecture (highlighted areas in green) at P5 (B–E) and P14 (G,H). Note the close spatial association of some myonuclei (highlighted areas in pink) with areas of myofiber lesions. (E,H) Potential influence of myonuclear domains in the spatial distribution of myofiber lesions: in the same myofiber segment coexist peripheral myonuclei (highlighted in pink) associated with areas of sarcomere disruption whereas, in the opposite side, other myonuclei (highlighted area in blue) associated with properly structured sarcomeres. Scale bars: 5 µm (A–H).

Figure 5

(A–D) Detail at high magnification of cytoskeletal alterations in SMN∆7 myofibers at P5. (A,B) Disassembly of the sarcomere architecture with loss of the banding pattern and disruption of Z-discs (highlighted areas). (C,D) In green, disarrayed myosin thick myofilaments (D) and aberrant accumulations of actin thin myofilaments are highlighted (F). (E–H) Immunogold electron microscopy analysis for the detection of myosin in myofibrils from WT (E) and SMN∆7 (F–H) mice at P5. (E) Gold particles of myosin immunoreactivity specifically decorate thick myofilaments in the properly aligned A-bands of WT sarcomeres. (F–H) In SMN∆7 myofibers there was misalignment of A-band (F) and disarray and loss of myosin-labeled thick myofilaments in sarcoplasmic areas of unstructured sarcomeres (G,H, highlighted areas in green). In (A,B) and (E–H), “Z” indicates Z-discs; “A” the A-band and “I” the hemi I-band. Scale bars: 2 µm (A); 1 µm (B,E–H); and 500 nm (C,D).

Figure 6

Electron micrographs of minisarcomeres in a SMN∆7 myofiber at P5. (A) Panoramic view of the longitudinal section of a myofiber illustrating a segment with shortened minisarcomeres juxtaposed to another segment with normal sarcomeres. Note the sharp transition between these two myofiber segments and the swelling of some mitochondria. (B,C) Detail of the comparative ultrastructure between normal sarcomeres from a WT myofiber (B), with the A-band flanked by two hemi-I-bands, and shortened minisarcomeres from a SMN∆7 myofiber (C), with the absence of I-bands and thick myofilaments that are anchored directly into Z-disc. In B-C, “Z” indicates Z-discs; “A” the A-band; “I” the hemi I-band and “M” the myosin thick myofilaments. Scale bars: 5 µm (A) and 1 µm (B,C).

Figure 7

(A–J) Ultrastructural organization of SR and triads in WT (A,F) and SMN∆7 (B–J,G–J) mice myofibers at P5. (A) Typical structure of triads, composed of a T-tubule flanked by two terminal SR cisterns (arrows), and intermyofibrillar longitudinal tubules of the SR network. (B,C) Dilation of SR cisterns associated with mitochondria (asterisks) and with the triads (arrows). Note the well-preserved fine structure of mitochondria. (D,E) Representative electron microscopy images of vacuolar degeneration affecting intermyofibrillar SR cisterns (asterisks). Note in panel D that some vacuoles directly interact with mitochondria. (F,G) Transversal sections of WT (F) and SMN∆7 (G) myofibers illustrating cross-sectioned myofibrils surrounded by a network of SR tubules in the WT sample and the presence of numerous dilated cisternae of SR in the SMN∆7 myofiber. (H–J) High magnification electron micrographs of sarcoplasmic areas of sarcomere disruption illustrating the disarray of SR tubules (asterisks) and triads (arrows). Scale bars: 1 µm (A–E,J), 5 µm (F,G), and 500 nm (H,I).

Figure 8

(A,B) Cross-cryosections of TA muscle stained with Mitotracker to analyze the mitochondrial content in WT and SMN∆7 myofibers at P5. The asterisk indicates a fast glycolytic type II myofibers. (C) Determination of the mean fluorescence intensity of MitoTracker signal per myofiber, using confocal images obtained from cross-cryosections of TA muscle from WT (n = 3) and SMN∆7 (n = 3) at P5 stained with Mitotracker. p value from WT and SMN∆7 data comparison was 0.0034. (D–F) Representative confocal mosaic images of TA muscle mid-belly transversal sections from WT and SMNΔ7 mice at P5, double immunostained using antibodies against myosin heavy chain I (MyHC-I, green), for slow-type I myofibers, and laminin (Lam, red), for myofiber contour visualization. Note the non-uniform distribution of MyHC-I positive myofibers along the TA muscle. (D’,E’) Higher magnification views of immunolabeled TA myofibers of the different experimental conditions. (F) Mean percentage of MyHC-I positive fibers in TA muscle from WT and SMNΔ7 mice at P5. All myofibers of a mid-belly transversal section of TA muscle from 4 animals per experimental condition were analyzed. p value from WT and SMN∆7 data comparison was 0.9023. In all graphs (C,F), values are shown as mean ± SEM, and unpaired Student’s t test (C) and two-way analysis of variance (Bonferroni’s post hoc test) (F) was used for statistical analysis; **: p < 0.005. Scale bar: 10 μm (A,B) 250 μm (D,E), and 25 μm (D’,E’).

Figure 9

(A,B) Electron micrographs of cross-sectioned myofibers of TA muscle from WT (A) and SMN∆7 (B) mice at P5. Note the accumulation of subsarcolemmal mitochondria in the SMN∆7 myofiber. (C,D) Quantitative analysis on electron micrographs from transversal TA myofiber sections from WT and SMNΔ7 mice at P5. The mean mitochondrial area (C) and the sarcoplasmic area occupied by mitochondria (D) were measured from at least 50 myofibers of each genotype using the Image J software (v.13.0.6. National Institute of Health, USA). p values from WT and SMN∆7 data comparison were 0.016 for the mean transversal mitochondrial area and 0.00015 for the mean sarcoplasmic area occupied by mitochondria. (E–H) qRT-PCR analysis of the expression levels of PCG1α, Fndc5, Mfn1, and Mfn2 in TA muscle samples from WT (n = 3) and SMN∆7 mice (n = 5) at P5. Note the significant increase of these gene transcripts in SMN∆7 RNA extracts relative to age matched WT samples. p values from WT and SMN∆7 data comparison were 0.0041 for PCG1α, 0.0316 for Fndc5, 0.0428 for Mfn1, and 0.0065 for Mfn2. (I,L) Ultrastructural changes in the mitochondrial phenotype of TA myofibers from SMN∆7 mice at P5. Note that the fine structure of mitochondria in WT myofibers at P5 is shown in Supplementary Figure S2A. (I,K) Presence of very long rod-like intermyofibrillar mitochondria in SMA myofibers. (J,L) Clusters of subsarcolemmal (J) and intermyofibrillar (L) mitochondria illustrating the interactome (organelle contact) between mitochondria as well as between mitochondria and lipid droplets (LD). Arrows indicate potential sites of mitochondrial fusion. In all graphs (F–H), values are shown as mean ± SD, and unpaired Student’s t test was used for statistical analysis; *: p < 0.05, **: p < 0.005, ***: p < 0.0005. Scale bar: 5 μm (A,B) and 2 μm (I–L).

Figure 10

(A–C) Representative confocal images of transversal (A,B,D) and longitudinal (C) cryosections from TA myofibers double or triple stained to label mitochondria (Mitotracker, red channel), thin actin filaments (Phalloidin, green channel) and nuclei (DAPI, blue channel) in WT (A) and SMN∆7 (B–D) samples at PND (P5). (A) Typical WT myofibers with peripheral myonuclei and intermyofibrillar and peripheral mitochondria (red spots). (B–D) Presence of some central myonuclei (B) and mitochondria-free bright green, fluorescent foci of F-actin accumulations (asterisks) in SMN∆7 myofibers. (E,F) Electron microscopy images of myofibers from SMN∆7 mice at PND stage (P5) showing extensive sarcoplasmic areas of disarrayed myofilaments and unrecognized sarcomeres free of mitochondria (highlighted green areas). Scale bar: 10 μm (A–D) and 5 μm (E,F).