Selective vulnerability in neuronal populations in nmd/SMARD1 mice

Abstract

Spinal muscular atrophy with respiratory distress type 1 (SMARD1) is an autosomal recessive motor neuron disease causing distal limb muscle atrophy that progresses proximally and is accompanied by diaphragmatic paralysis. Neuromuscular junction (NMJ) alterations have been reported in muscles of SMARD1 model mice, known as nmd mice, with varying degrees of severity, suggesting that different muscles are specifically and selectively resistant or susceptible to denervation. To evaluate the extent of NMJ pathology in a broad range of muscles, a panel of axial and appendicular muscles were isolated and immunostained from nmd mice. These analyses revealed that selective distal appendage muscles were highly vulnerable to denervation. Susceptibility to pathology was not limited to NMJ alterations, but included defects in myelination within those neurons innervating susceptible muscles. Interestingly, end plate fragmentation was present within all muscles independent of the extent of NMJ alterations, suggesting that end plate fragmentation is an early hallmark of SMARD1 pathogenesis. Expressing the full-length IGHMBP2 cDNA using an adeno-associated virus (AAV9) significantly decreased all aspects of muscle and nerve disease pathology. These results shed new light onto the pathogenesis of SMARD1 by identifying specific motor units that are resistant and susceptible to neurodegeneration in an important model of SMARD1.

Figure 1.

Vulnerability to NMJ denervation is more pronounced in distal appendicular muscles and can be prevented by AAV9-IGHMBP2 treatment. Quantification of NMJ pathology showing percentages of fully innervated, partially innervated, and fully denervated end plates from 8-week-old wild type, nmd, and nmd + AAV9-IGHMBP2 treated mice. (A) Quantification of NMJ pathology in resistant (cLAL and masseter) and slightly vulnerable (LC and SC) muscles from the neck. (B) Quantification of NMJ pathology in resistant (diaphragm and SPI) and vulnerable (LD and TVA) muscles from the trunk. (C) Quantification of NMJ pathology in vulnerable (gastrocnemius, TA, FDB-2, FDB-3, and FDB-4) and resistant (EDL and lumbricals) muscles from distal appendages. Quantification of NMJ pathology also shows that AAV9-IGHMBP2 treatment of nmd mice significantly reduces NMJ pathology in all muscle groups analyzed (A, B, and C). Data were analyzed by one-way ANOVA followed by a Bonferroni post hoc test for multiple comparisons. Data expressed as mean ± SEM. *P < 0.05, **P < 0.001, and ***P < 0.0001. n.s., not significant. n = 3 animals per treatment.

Figure 2.

Increased vulnerability to NMJ denervation in distal appendicular muscles and prevention with AAV9-IGHMBP2 treatment. Neck (anterior), trunk (middle), and distal appendicular muscles (posterior) were immunostained to label axon (NF-H), axon terminal (SV2), and end plate (AChRs). (Top panel) Images of representative muscle (cLAL) from the neck showing resistance to NMJ denervation in nmd (middle) and nmd+ AAV9-IGHMBP2 treated (right) compared to an unaffected/wild-type control (left) (scale bars = 100 µm). (Middle panel) Images of a representative resistant muscle (diaphragm) from the trunk area in nmd (middle) and nmd+ AAV9-IGHMBP2 treated (right) compared to an unaffected control (left) (scale bars = 30 µm). (Bottom two panels) Representative muscles (gastrocnemius and FDB-2) from distal appendicular area with high vulnerability to NMJ denervation in nmd (middle) compared to control (left) (scale bars = 30 µm). AAV9-IGHMPB2 treatment successfully prevents NMJ denervation in highly vulnerable muscles (bottom right). Maximum projection confocal microscope images taken at 20× magnification.

Figure 3.

Differential vulnerability to denervation predominantly affects neurons of distal muscles. Quantification of immunostained muscles and comparison of percent available fully innervated end plates across all muscles analyzed in 8-week wild-type and nmd mice. Individual comparisons were analyzed by Student’s t-test. Data are expressed as mean ± SEM. *P < 0.05. n = 3 animals per treatment.

Figure 4.

End plate fragmentation is present in nmd muscles regardless of resistance to NMJ denervation. (A) Quantification of percent normal and fragmented end plates in muscles from 8-week wild-type, nmd, and nmd+AAV9-IGHMBP2 treated mice. End plate fragmentation in various selected muscles with differential vulnerability to NMJ denervation encompassing neck, trunk, and distal appendages. AAV9-IGHMPB2 treatment of nmd mice prevents end plate fragmentation in all muscles analyzed. (B) Representative images of immunostained muscle showing presence of end plate fragmentation (arrows) in a muscle resistant to NMJ denervation (cLAL) and prevention of end plate fragmentation by AAV9-IGHMBP2 treatment. Data were analyzed by one-way ANOVA followed by a Bonferroni post hoc test for multiple comparisons. Data are expressed as mean ± SEM. **P < 0.001; ***P < 0.0001. n.s., not significant. n = 3 animals per treatment.

Figure 5.

Tibial nerve fibers show demyelination in nmd mice. (A) (Left panel) Representative sample of wild-type tibial nerve fibers indicating location of Schwann cell nucleus (star) and nodes of Ranvier (arrows). (Right panel) High magnification images showing wild-type axons with normal myelin, node of Ranvier (arrows), and Schwann cell nucleus (star). (B) (Left panel) Representative sample of nmd tibial nerve teased fibers showing a bundle of axons completely demyelinated (lightning bolt), axons with myelin fragmentation (triangle) and axons with normal myelin and nodes of Ranvier (arrows). Note that in demyelinated axons and fragmented myelin axons, no nodes of Ranvier can be identified. (Left panel) High magnification images of axons with fragmented myelin.

Figure 6.

Pathology in tibial nerve from nmd mice. (A) Scatter plots showing the correlation between internode length and fiber diameter of nmd and wild type controls. r² values of regression correlations showed a significant shift towards smaller fiber diameter with shorter internode lengths in the population of nmd fibers compared with wild-type controls (P < 0.0001). (B) Mean fiber diameter comparisons revealed a significant (P = 0.0001) decrease in nmd fiber diameter (6.19 ± 0.07 µm) compared to wild-type fiber diameter (7.89 ± 0.09 µm). (C) Mean internode length comparisons revealed a significant (P < 0.0001) decrease in nmd (554.8 ± 5.69 µm) compared to wild-type (707.6 ± 6.1 µm) internode length. (D) Quantification of abundance of fibers with myelin alterations (myelin fragmentation, segmental demyelination, complete demyelination) revealed a significant increase (0 = 0.0135) in percent of total fibers teased with myelin defects in nmd (37%) compared to wild-type controls (0.6%). Individual comparisons were analyzed by Student’s t-test. Data expressed as mean ± SEM. *P < 0.05. n = 3 animals per treatment.

Figure 7.

No pathology was found in nerve innervating the masseter muscle. (A) Scatter plot showing correlations of internode length versus fiber diameter correlations in nmd and wild-type controls. No differences were found in r² regression correlation values between wild-type and nmd mice. (B) Mean fiber diameter comparisons between wild-type and nmd mice showed no statistical differences (P = 0.095). (C) Comparisons of mean internode length also revealed no differences (P = 0.123) between wild-type and nmd mice. (D) Quantification of abundance of fibers with myelin defects revealed no difference (P = 0.528) between wild-type and nmd mice. Individual comparisons were analyzed by Student’s t-test. Data are expressed as mean ± SEM. n = 3 animals per treatment.