Depletion of SMN Protein in Mesenchymal Progenitors Impairs the Development of Bone and Neuromuscular Junction in Spinal Muscular Atrophy

Sang-Hyeon Hann
Seon-Yong Kim
Ye Lynne Kim
Young-Woo Jo
Jong-Seol Kang
Hyerim Park
Se-Young Choi author has email address
Young-Yun Kong author has email address

School of Biological Sciences, Seoul National University, Seoul 08826, Republic of Korea
Department of Physiology, Dental Research Institute, Seoul National University School of Dentistry, Seoul, 03080, Republic of Korea

https://doi.org/10.7554/eLife.92731.2

Open access
Copyright information

Figures and data

Skeletal growth abnormalities and altered growth plate homeostasis in SMN2 1-copy Smn^ΔMPC mice.
(A) Representative 3D images and longitudinal section view of the ossified femur bone. Scale bars, 1 mm. (B-C) SMN2 1-copy mutant’s femurs showed reduced growth in diaphysis length and diameter, and (D) decreased trabecular bone volume. (E) Trabecular bone thicknesses were not significantly different between the control and mutant groups. The micro CT analysis was performed in femur diaphysis and metaphysis from SMN2 1-copy Smn^ΔWT, SMN2 2-copy, and 1-copy Smn^ΔMPC mice at P14. 1-way ANOVA with Tukey’s post hoc test, n = 3-5 mice in each genotype (B-E). (F) Representative images of H&E staining in the distal femur growth plate of control and mutant mice with 1 copy of SMN2 at P14. Scale bars, 100 μm. Resting zone (RZ), Hypertrophic zone (HZ) and proliferative zone (PZ). (G-I) Indicated by black arrows, the HZ and PZ lengths were reduced in SMN2 1-copy Smn^ΔMPC mice, and the hypertrophic cell number in a section of the 1-copy mutant was decreased (n = 4 mice in each genotype; Unpaired t-test with Welch’s correction). (J) Representative images of Ki67 immunostaining in the distal femur growth plate of control and mutant mice with 1 copy of SMN2 at P14. Scale bars, 100 μm. (K) and decreased Ki67+ percentage in resting zone chondrocytes. (L) Ki67+ percentage in the proliferative zone was not significantly different between the control and mutant groups. n = 3 mice in each genotype; Unpaired t-test with Welch’s correction (K-L). ns; not significantly different. *p < 0.05; **p < 0.01. Error bars show s.e.m.

Decreased chondrocytes-derived IGF-AKT axis by limb mesenchymal cell-specific SMN depletion in SMN2 1-copy Smn^ΔMPC mice.
(A) Representative images of p-AKT immunostaining in distal femur growth plate from mice at P14. Scale bars, 100 μm. (B-C) The p-AKT-positive percentage was decreased in the resting zone chondrocytes, not in the proliferative zone. (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction) (D) Relative IGF axis mRNA expression in the livers of SMN2 1-copy control and mutant mice. The IGF pathway genes showed no difference when comparing controls to SMN2 1-copy Smn^ΔMPC mutants. (E-F) Relative IGF axis and chondrocyte differentiation marker mRNA expression in the chondrocytes of SMN2 1-copy control and mutant mice. The Igf1, Igfbp3, and hypertrophic marker Col10a1 expression were decreased in SMN2 1-copy Smn^ΔMPC mutants. n = 3 mice in each genotype; Unpaired t-test with Welch’s correction (D-F). (G) Representative images of genomic PCR analysis from SMN2 1-copy Smn^ΔMPC mice tissues at P21. (H) qRT PCR analysis from tissues of SMN2 1-copy Smn^WT and Smn^ΔMPC mice at P21 (n = 3 mice in each genotype; Unpaired t-test with Welch’s correction). Deletion of Smn exon 7 was detected only in limb mesenchymal cells using genomic PCR (G) and full-length SMN mRNA expression (H). (I) Representative images of western blot analysis in cultured FAPs. SMN protein in FAPs of SMN2 1-copy Smn^ΔMPC mice exhibited a decrease comparable to that observed in the SMAΔ7 mice. (J) Relative SMN levels in cultured FAPs of the controls and mutants (n = 3 mice in each genotype; 1-way ANOVA with Tukey’s post hoc test). ns; not significantly different. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Error bars show s.e.m.

Aberrant postnatal NMJ maturation in SMN2 1-copy Smn^ΔMPC mice.
(A) Immunostaining of NMJs in TA muscle of SMN2 1-copy Smn^WT and Smn^ΔMPC mice at P21 with anti-NF (Green), anti-synaptophysin (Blue), α-Btx staining AChR (Red). Scale bars, 20 μm. The confocal images of NMJs showed decreased presynaptic terminal branching and the existence of nerve terminal varicosities that were enlarged with NF (indicated by arrowheads) in the mutant. (B) The NMJs of the SMN2 1-copy mutant exhibited a significant decrease in presynaptic terminal arborization and (C) an increased percentage of poorly arborized NMJs (n = 3 mice in each genotype; Unpaired t-test with Welch’s correction). (D) The percentage of NMJs exhibiting NF varicosities was higher in the SMN2 1-copy mutant group than in the control group (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction). (E) For quantification of the NMJ maturation stage, we classified NMJs into five distinct developmental stages (Plaque: Plaque-shaped endplate without any perforation; Small perforated: Plaque-shaped endplate with small perforations; Large perforated: Plaque-shaped endplate with large perforations; Open: C-shaped endplate; Pretzel: Pretzel-like shaped endplate) and then compared the frequency patterns of SMN2 1-copy control and mutant mice (n = 3 mice in each genotype; 2-way ANOVA with Tukey’s post hoc test). The NMJs of SMN2 1-copy mutants displayed plaque-like shapes, indicating that they were in the immature stage. (F) Immunostaining of NMJs in TA muscle of SMN2 1-copy Smn^WT and Smn^ΔMPC mice at P3 with anti-NF (Green), anti-synaptophysin (Blue), α-Btx staining AChR (Red). Scale bars, 10 μm. (G) There were no significant differences in AChR cluster size between the SMN2 1-copy control and mutant at P3 (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction). (H) The ratio of the Synaptophysin area to the AChR area in NMJ was slightly higher in the SMN2 1-copy mutant at P3 (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction). ns; not significantly different. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. All box-and-whisker plots show the median, interquartile range, minimum, and maximum. For the box-and-whisker plots, range bars show minimum and maximum (B, G, H). For the bar and line graph, error bars show s.e.m(C, D, E).

Morphological deterioration in NMJs of adult SMN2 1-copy Smn^ΔMPC mice.
(A) Immunostaining of NMJs in TA muscle of SMN2 1-copy Smn^WT, SMN 2-copy, and SMN2 1-copy Smn^ΔMPC mice at P56 with anti-NF (Green), anti-synaptophysin (Blue), α-Btx staining AChR (Red). Scale bars, 20 μm. The confocal images of NMJs showed fragmentation and bouton-like NF varicosities (indicated by arrowheads) in the SMN2 1-copy Smn^ΔMPC mice. (B-D) The NMJs of the SMN2 1-copy mutant displayed fragmented presynapse (B), endplate (D), and NF varicosities (C) compared to SMN2 1-copy Smn^WT and SMN2 2-copy Smn^ΔMPC mice (n = 3-5 mice in each genotype; Presynaptic fragments & AChR fragments: Brown– Forsythe and Welch ANOVA with Games-Howell’s test; NF varicosities: 1-way ANOVA with Tukey’s post hoc test). ns; not significantly different. ****p < 0.0001. All box-and-whisker plots show the median, interquartile range, minimum, and maximum. For the box-and-whisker plots, range bars show minimum and maximum (B, D). For the bar graph, error bars show s.e.m. (C)

Reduced presynaptic neurotransmission ability in the NMJs of SMN2 1-copy Smn^ΔMPC mice.
(A) Representative traces of mEPP from SMN2 1-copy Smn^ΔWT (top) and SMN2 1-copy Smn^ΔMPC (bottom) mice. (B-C) SMN2 1-copy mutant’s NMJs showed an increase in mEPP amplitude and no differences in mEPP frequency (1-copy control, n = 25, 9 mice; 1-copy mutant, n = 21, 8 mice; Unpaired t-test with Welch’s correction). (D) Representative traces of eEPP from SMN2 1-copy Smn^ΔWT (top) and SMN2 1-copy Smn^ΔMPC (bottom) mice. (E) The mutant’s NMJs showed a stronger amplitude of eEPPs (1-copy control, n = 12, 4 mice; 1-copy mutant, n = 12, 3 mice; Unpaired t-test with Welch’s correction). (F) Representative traces of paired-pulse response from SMN2 1-copy Smn^ΔWT (top) and SMN2 1-copy Smn^ΔMPC (bottom) mice. (G) Paired-pulse response was not different between SMN2 1-copy control and mutant NMJs, indicating a comparable neurotransmitter release probability (1-copy control, n = 8, 3 mice; 1-copy mutant, n = 6, 3 mice; Unpaired t-test with Welch’s correction). The electrophysiological recording was performed in the EDL muscle at P56. ns; not significantly different. *p < 0.05. All box-and-whisker plots show the median, interquartile range, minimum, and maximum. For the box-and-whisker plots, range bars show minimum and maximum (B, C, E, G).

Abnormal nerve terminal ultrastructure in SMN2 1-copy mutant.
(A) Representative TEM images from NMJs of SMN2 1-copy Smn^ΔWT and SMN2 1-copy Smn^ΔMPC mice at P56. Scale bars, 500 nm. Nerve terminal (NT; indicated by the blue zone). Synaptic vesicles (SVs). Muscle fiber (MS; indicated by the red zone). Endplate junctional folds (JF). Nerve terminal detachment (indicated by arrow) was observed in SMN2 1-copy Smn^ΔMPC mice. (B) The density of junctional folds in the NMJ of SMN2 1-copy Smn^ΔMPC mice no significant change compared to the control, whereas (C) the density of synaptic vesicles was increased (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction). (D) The detachment of the nerve terminal occurs more frequently at the NMJ of mutants (n = 3-4 mice in each genotype; Unpaired t-test with Welch’s correction). ns; not significantly different. *p < 0.05; ***p < 0.001. All box-and-whisker plots show the median, interquartile range, minimum, and maximum. For the box-and-whisker plots, range bars show minimum and maximum (B, C). For the bar graph, error bars show s.e.m(D).

Improved postnatal NMJ development in the TA muscle of SMN2 1-copy Smn^ΔMPC mice following healthy FAPs transplantation.
(A) Immunostaining of NMJs in tdTomato⁺ FAPs-transplanted TA muscle (+FAP) and vehicle-treated contralateral muscle (+Veh) in SMN2 1-copy Smn^ΔMPC mice at P56 with anti-NF (Yellow), anti-synaptophysin (Blue), α-Btx staining AChR (Green), and tdTomato fluorescence (Red). Scale bars, 40 μm. The images revealed NF varicosities (indicated by arrowheads) in the +Veh NMJs. The tdTomato+ FAPs (marked by asterisks) were transplanted into +FAP NMJs, which exhibited (B) decreased presynaptic fragmentation, (C) NF varicosities, and (D) AChR fragmentation compared to +Veh NMJs and similar to wild-type NMJs. (n = 3-4 mice in each group; Unpaired t-test with Welch’s correction). ns; not significantly different. **p < 0.01, ***p < 0.001, ****p < 0.0001. All box-and-whisker plots show the median, interquartile range, minimum, and maximum. For the box-and-whisker plots, range bars show minimum and maximum (B, D). For the bar graph, error bars show s.e.m (C).

Growth defects in the Prrx1-lineage bone of SMN2 0-copy Smn^ΔMPC mice.
(A) Representative control and mutant mice with 0 copies of SMN2 were photographed at E18.5. (B-C) Alcian blue & alizarine red staining in the SMN2 0-copy Smn^ΔMPC mutant showed abnormal skeletal development of the limbs, (D) calvaria (shown by the dashed line), and (E) sternum (indicated by a white arrow). Scale bars, 5 mm (B-C) and 2 mm (D-E).

Osteoclasts and osteoblasts were undisturbed in SMN2 1-copy Smn^ΔMPC mice.
(A) Cortical bone thicknesses were not significantly different between the control and mutant groups. (B) Bone mineral density was slightly increased in the diaphysis of SMN2 1-copy Smn^ΔMPC mice. The micro CT analysis was performed in femur diaphysis and metaphysis from SMN2 1-copy Smn^WT, SMN2 2-copy, and 1-copy Smn^ΔMPC mice at P14. 1-way ANOVA with Tukey’s post hoc test, n = 3-5 mice in each genotype (A-B). (C) Representative images of osteoclast marker Itgb3 immunostaining in femur diaphysis cortical bone from mice at P14. Scale bars, 100 μm. Itgb3+ hematopoietic cell (indicated by arrow) and osteoclast (indicated by arrowhead) were imaged. (D) It showed similar osteoclast density. n = 3 mice in each genotype; Unpaired t-test with Welch’s correction. (E) Representative images of toluidine blue staining in femur diaphysis cortical bone for osteoblast evaluation. Scale bars, 100 μm. ns; not significantly different. *p < 0.05. Error bars show s.e.m.

Sign up for email alerts