MBNL deficiency in motor neurons disrupts neuromuscular junction maintenance and gait coordination

Abstract

Muscleblind-like proteins (MBNLs) are a family of RNA-binding proteins that play essential roles in the regulation of RNA metabolism. Beyond their canonical role in RNA regulation, MBNL proteins have emerged as key players in the pathogenesis of myotonic dystrophy type 1. In myotonic dystrophy type 1, sequestration of MBNL proteins by expansion of the CUG repeat RNA leads to functional depletion of MBNL, resulting in deregulated alternative splicing and aberrant RNA processing, which underlie the clinical features of the disease. Although attention on MBNL proteins has focused on their functions in skeletal muscle, new evidence suggests that their importance extends to motor neurons (MNs), pivotal cellular components in the control of motor skills and movement. To address this question, we generated conditional double-knockout (dKO) mice, in which Mbnl1 and Mbnl2 were specifically deleted in motor neurons (MN-dKO). Adult MN-dKO mice develop gait coordination deficits associated with structural and ultrastructural defects in the neuromuscular junction, indicating that MBNL activity in MNs is crucial for the maintenance of the neuromuscular junction. In addition, transcriptome analysis performed on the spinal cord of MN-dKO mice identified mis-splicing events in genes associated with synaptic transmission and neuromuscular junction homeostasis. In summary, our results highlight the complex roles and regulatory mechanisms of MBNL proteins in MNs for muscle function and locomotion. This work provides valuable insights into fundamental aspects of RNA biology and offers promising avenues for therapeutic intervention in myotonic dystrophy type 1 and in a range of diseases associated with RNA dysregulation.

Keywords: MBNL; RNA binding protein; alternative splicing; motor neurons; myotonic dystrophy type 1; neuromuscular junction.

Figure 1

MBNL1 and MBNL2 combined depletion in motor neurons affects mouse coordination. (A) Schematic representation of the breeding strategy used to deplete MBNL2 specifically in motor neurons (MNs) in transgenic mice. Mbnl1^+/−; Mbnl2^LoxP/LoxP mice were crossed with Mxn1Cre mice (also known as Hb9Cre mice) to generate: Mbnl1^+/+; Hb9-Mbnl2^+/+ control mice (CTL); Mbnl1^−/−; Hb9-Mbnl2^−/− MN-dKO mice; Mbnl1^−/−; Hb9-Mbnl2^+/+  Mbnl1-KO mice; and Mbnl1^+/+; Hb9-Mbnl2^−/− Hb9Cre-Mbnl2-KO mice. (B) Representative trace of footprint analysis for testing gait abnormalities in CTL, HB9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice at 2 months of age. (C–G) Quantification from the gait analyses for 2-month-old CTL, HB9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice. (C) Proportions of walking patterns, defined by the limbs mice are standing on at the same time during walking. ‘Others’ refers to standing on three or four limbs, and alternative walking patterns. (D) Ratio of the mean intensity of pixels corresponding to front leg pawprints on the mean intensity of pixels corresponding to hind leg pawprints. (E) Quantification of the mean stride length of all paws. (F) Quantification of the mean standing time of all paws. (G) Quantification of front/hind paw distance of either ipsilateral combination of paws. Data are mean ± standard error of the mean from 2-month-old n_CTL = 14; n_{Hb9Cre-Mbnl2-KO} = 8; n_Mbnl1-KO = 11; n_MN-dKO = 8 mice. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001; Tukey's test. dKO = double knockout; MN = motor neuron.

Figure 2

MBNL1 and MBNL2 combined depletion in motor neurons leads to early mouse muscle strength reduction. Analysis of tibialis anterior muscles from 2-month-old control (CTL), Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice. (A) Haematoxylin and eosin staining of tibialis anterior cryosections. Scale bar = 100 μm. (B) in situ measurement of myotonia (in grams seconds). (C) Tibialis anterior in situ muscle maximal specific force (in grams per milligram). Data are mean ± standard error of the mean from 2-month-old mice: n_CTL = 4; n_{Hb9Cre-Mbnl2-KO} = 3; n_Mbnl1-KO = 4; n_MN-dKO = 4 mice. *P < 0.05; one-way ANOVA, Tukey's test. dKO = double knockout; MN = motor neuron.

Figure 3

Progressive worsening of Mbnl1-KO and MN-dKO locomotor abilities. (A) Representative trace of footprint analysis for testing gait abnormalities in control (CTL), Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice at 4 months of age. (B–F) Quantification from the gait analyses for 2-month-old CTL, HB9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice. (B) Proportions of walking patterns, defined by the limbs mice are standing on at the same time during walking. ‘Others’ refers to standing on three or four limbs, and alternative walking patterns. (C) Ratio of the mean intensity of pixels corresponding to front leg pawprints on the mean intensity of pixels corresponding to hind leg pawprints. (D) Quantification of the mean stride length of all paws. (E) Quantification of the mean standing time of all paws. (F) Quantification of front/hind paw distance of either ipsilateral combination of paws. Data are mean ± standard error of the mean from 4-month-old mice: n_CTL = 11; n_{Hb9Cre-Mbnl2-KO} = 6; n_Mbnl1-KO = 7; n_MN-dKO = 5 mice. *P < 0.05, **P < 0.01; one-way ANOVA, Dunn's test (B, C and F); one-way ANOVA, Tukey's test (D and E). dKO = double knockout; MN = motor neuron.

Figure 4

Progressive muscle histology and strength modifications in Mbnl1-KO and MN-dKO mice. Analysis of tibialis anterior muscles from 4-month-old control (CTL), Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice. (A) Haematoxylin and eosin staining of tibialis anterior cryosections. White arrows = split fibres; black arrows = centronucleated fibres. Scale bar = 100 μm. (B) Proportions of centronucleated fibres in tibialis anterior muscle transverse cryosections. (C) In situ measurement of myotonia (in grams seconds). (D) Tibialis anterior in situ muscle maximal specific force (in grams per milligram). Data are mean ± standard error of the mean from 4-month-old mice: n_CTL = 7; n_{Hb9Cre-Mbnl2-KO} = 6; n_Mbnl1-KO = 7; n_MN-dKO = 7 mice (B); and n_CTL = 14; n_{Hb9Cre-Mbnl2-KO} = 12; n_Mbnl1-KO = 11; n_MN-dKO = 8 mice (C and D). *P < 0.05, ***P < 0.001 and ****P < 0.0001; one-way ANOVA, Tukey's test. dKO = double knockout; MN = motor neuron.

Figure 5

MN-dKO mice exhibit neuromuscular junction structural alterations at 2 months of age. (A) Representative confocal images of isolated muscle fibres from 2-month-old control (CTL), Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO tibialis anterior muscle stained with antibodies to neurofilament (NF) and synaptophysin (Syn, magenta) together with α-bungarotoxin (α-BTX; AChR, green), with quantitative analysis in B–G. (B) AChR cluster area. (C) Neuromuscular junction (NMJ) nerve terminal area. (D) End-plate perimeter. (E) Index of AChR fragmentation. (F) Percentages of NMJs that contain indicated numbers of AChR fragments. (G) Overlap ratio of pre- and postsynaptic AChR cluster area staining. Data are mean ± standard error of the mean from n_CTL = 4; n_{Hb9Cre-Mbnl2-KO} = 3; n_Mbnl1-KO = 3; n_MN-dKO = 4. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001; one-way ANOVA, Dunn's test (B, E and G); one-way ANOVA, Tukey's test (C and D); and two-way ANOVA, Tukey's test (F). dKO = double knockout; MN = motor neuron.

Figure 6

Aggravation of neuromuscular junction structural alterations in Mbnl1-KO and MN-dKO mice with age. (A) Representative confocal images of isolated muscle fibres from 4-month-old control (CTL), Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO tibialis anterior muscle stained with antibodies to neurofilament (NF) and synaptophysin (Syn, magenta) together with α-bungarotoxin (α-BTX; AChR, green), with quantitative analysis in B–H. (B) AChR cluster area. (C) Neuromuscular junction (NMJ) nerve terminal area (D) End-plate perimeter. (E) Index of AChR fragmentation. (F) Percentages of NMJs that contain indicated numbers of AChR fragments. (G) Overlap ratio of pre- and postsynaptic AChR cluster area staining. (H) Proportion of NMJs with nerve terminal sprouts (indicated in A with a white arrow and magnified in the left bottom corner). Data are mean ± standard error of the mean from n_CTL = 4; n_{Hb9Cre-Mbnl2-KO} = 7; n_Mbnl1-KO = 7; n_MN-dKO = 6. *P < 0.05, **P < 0.01 and ***P < 0.001; one-way ANOVA, Tukey's test (H); one-way ANOVA, Dunn's test (B–E and G); and two-way ANOVA, Tukey's test (F). dKO = double knockout; MN = motor neuron.

Figure 7

Additional loss of MBNL2 in motor neurons leads to neuromuscular junction ultrastructural modifications with presynaptic defects. Representative electron micrograph of neuromuscular junctions from 4-month-old control (CTL), Mbnl1-KO and MN-dKO tibialis anterior. Synaptic boutons are coloured purple; black squares focus on a synapse magnified on the right. Degenerative structures are shown by arrows in the presynaptic (yellow) and postsynaptic (red) compartments. dKO = double knockout; MN = motor neuron.

Figure 8

Alternative splicing changes in spinal cord from 4-month-old MN-dKO mice. (A) Pie chart represents the proportion of significant differential alternative splicing (DSE) (ΔΨ > 10%, false discovery rate < 0.05) in 4-month-old MN-dKO mice compared with control (CTL) mice and categorized as skipped exons (SE), alternative 5′ and 3′ splice sites (A5SS and A3SS), mutually exclusive exons (MXE) and retained introns (RI). (B) Unbiased gene ontology (GO) top 15 biological processes involved by the differential alternative splicing events (DSE) identified in MN-dKO mice when compared with CTL mice with their corresponding P-values and enrichment ratio. (C) Heatmap shows selected DSE changed in MN-dKO mice compared with CTL mice. (D) Sashimi plot for Dvl1 ex14b in CTL and MN-dKO mice. (E) Representative RT-PCR splicing analysis of Dvl1 ex14b from CTL (n = 3), Mbnl1-KO (n = 3), Hb9Cre-Mbnl2-KO (n = 2) and MN-dKO mice (n = 3) with per cent spliced in (PSI) quantification. (F) Venn diagram of the number of unique genes affected in Hb9Cre-Mbnl2-KO, Mbnl1-KO and MN-dKO mice when compared with CTL mice. (G and H) The top 15 unbiased GO biological processes of genes common between MN-dKO and Mbnl1-KO mice (G) or Hb9Cre-Mbnl2-KO (H) mice are shown with their corresponding P-values and enrichment ratio. dKO = double knockout; MN = motor neuron.