MBNL-dependent impaired development within the neuromuscular system in myotonic dystrophy type 1

Abstract

Aims: Myotonic dystrophy type I (DM1) is one of the most frequent muscular dystrophies in adults. Although DM1 has long been considered mainly a muscle disorder, growing evidence suggests the involvement of peripheral nerves in the pathogenicity of DM1 raising the question of whether motoneurons (MNs) actively contribute to neuromuscular defects in DM1.

Methods: By using micropatterned 96-well plates as a coculture platform, we generated a functional neuromuscular model combining DM1 and muscleblind protein (MBNL) knock-out human-induced pluripotent stem cells-derived MNs and human healthy skeletal muscle cells.

Results: This approach led to the identification of presynaptic defects which affect the formation or stability of the neuromuscular junction at an early developmental stage. These neuropathological defects could be reproduced by the loss of RNA-binding MBNL proteins, whose loss of function in vivo is associated with muscular defects associated with DM1. These experiments indicate that the functional defects associated with MNs can be directly attributed to MBNL family proteins. Comparative transcriptomic analyses also revealed specific neuronal-related processes regulated by these proteins that are commonly misregulated in DM1.

Conclusions: Beyond the application to DM1, our approach to generating a robust and reliable human neuromuscular system should facilitate disease modelling studies and drug screening assays.

Keywords: MBNL proteins; Motoneurons; induced pluripotent stem cells; myotonic dystrophy type 1; neuromuscular junction.

FIGURE 1

Myotonic dystrophy type I (DM1) human‐induced pluripotent stem cell (hiPSC)‐derived motoneurons (MNs) reproduce the main features associated with DM1. (A) Schematic representation of the differentiation protocol to convert hiPSC into MNs (created with BioRender.com ) for the different hiPSC lines listed in the table below. This summary table recapitulates the sex, the age of each patient, their main clinical features and associated MDRS (muscular disability rating scale) at the time the skin fibroblast biopsies were performed. Fibroblasts from three DM1 patients and one healthy donor subject (control [CTRL]; sibling of DM1_2 patient) were used. The CTG length determined by southern blot analysis in fibroblasts and hiPSC is indicated. On the right, a schematic representation of the generation of muscleblind protein (MBNL) knock‐out hiPS cell lines. (B) Representative images of immunostaining for motoneuronal markers in hiPSC‐derived MNs after 30 days of differentiation. MNs were identified by Islet1 (ISL1; Green) and TuJ1 (red) immunostaining. Nuclei were revealed by Hoechst staining (blue). Scale bar 10 μm. On the right, the quantification of the percentage of ISL1 and TuJ1‐positive hiPSC‐derived MNs. Data represent the mean ± SD values from at least three independent experiments in technical triplicate. Data were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with CTRL (ns: not significant). (C) Representative images of mutant DMPK mRNA foci (red) detected by mRNA fluorescence in situ hybridisation combined with immunolabelling for ISL1 (green) in CTRL, DM1 and DM1_ 3_ΔCTG‐depleted hiPSC‐derived MNs after 30 days of differentiation. Nuclei were detected by Hoechst staining. Scale bar 10 μm. The graph on the left represents the quantification of the percentage of ISL1‐positive cells containing foci. Data represent the mean ± SD values from at least three independent experiments in technical triplicate and were analysed with an ordinary one‐way ANOVA, Tukey's multiple comparisons test compared with CTRL (****: p ≤ 0.0001, ns: not significant).

FIGURE 2

Muscleblind protein (MBNL)‐dependent abnormal neuritic outgrowth in myotonic dystrophy type I (DM1) human‐induced pluripotent stem cell (hiPSC)‐derived MNs. (A) Quantification of neurite length (μm) per hiPSC‐derived motoneuron (MN) at day 30 of differentiation by co‐immunostaining for TuJ1 and Islet1 (ISL1). Data are represented as the mean ± SD values from at least four independent experiments in technical triplicate and were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with control (CTRL) (***: p ≤ 0.001, ****: p ≤ 0.0001, ns: not significant). (B) Percentage of GFP‐positive cells and ISL1‐positive cells expressing GFP following pAAV2[Exp]‐EF1A>EGFP:WPRE transduction of hiPSC‐derived MNs (at day 30). Data represent the mean ± SD values from three independent experiments in three technical triplicates and were analysed with Student's t‐test (ns: not significant). (C) Representative images of MBNL1 immunostaining (green, nuclei in blue) in CTRL, DM1, MBNL1 knock‐out and DM1 hiPSC‐derived MNs transduced by pAAV2[Exp]‐EF1A>hMBNL1[NM_021038.5]:WPRE. Scale bar 10 μm. MBNL1 overexpression was determined by the quantification of immunofluorescence in hiPSC‐derived MNs 20 days posttransduction with pAAV2[Exp]‐EF1A>hMBNL1[NM_021038.5]:WPRE. Data were normalised on the values generated for mock‐transduced DM1_2. Data are represented as the mean ± SD values from three independent experiments in technical triplicate and were analysed with an ordinary one‐way ANOVA, Tukey's multiple comparisons test compared with DM1_2 (***: p ≤ 0.001, *: p < 0.05). (D) Representative images of colabelling for ISL1 (green) and TuJ1 (red) in CTRL, DM1 and MBNL1 transduced hiPSC‐derived MNs. Scale bar 10 μm. Quantification of the neuritic outgrowth (μm) per hiPSC‐derived MNs transduced by pAAV2[Exp]‐EF1A>hMBNL1[NM_021038.5]:WPRE was determined at day 30 of differentiation. Data represent the mean ± SD values from three independent experiments in technical triplicate (highlighted with nuanced colours) and were analysed with an ordinary one‐way ANOVA, Tukey's multiple comparisons test compared with CTRL and DM1_2 (*: p < 0.05, **: P < 0.005, ns: not significant).

FIGURE 3

Human‐induced pluripotent stem cell (hiPSC)‐derived motoneurons (MNs) enhance acetylcholine receptor (AChR) clustering after 7 days of coculture. (A) Schematic representation of the protocol developed for the coculture between hiPSC‐derived MNs and human skeletal muscle cells. MNs differentiated from hiPSC for 14 days were dissociated and plated on top of human primary skeletal muscle cells differentiated for 2 days. Cocultures were kept at least for 7 days. (B) Representative images of an entire well from a 96‐well plate immunostained for myosin heavy chain (MF20, green) and TuJ1 (red). Nuclei were detected by Hoechst staining (blue). Each well contains about 100 micropatterns. Scale bar: 100 μm. On the right, a higher magnification of one micropattern stained for TuJ1 (Grey), acetylcholine receptors (AChR, red) and myosin heavy chain (MF20, Green). Nuclei were detected by Hoechst (blue). Scale Bar 10 μm. (C) Representative images of hiPSC‐derived MNs stained for ISL1 (Green), and TuJ1 (red). Nuclei were detected by Hoechst staining (blue). Scale bar 10 μm. On the right, quantification of the mean number of hiPSC‐derived MNs per pattern determined by ISL1 positive cells. Data represent the mean ± SD values from 7 independent experiments in technical triplicate (at least 100 micropatterns/condition/experiment). (D) Representative images of skeletal muscle cells immunolabelled using myosin heavy chain marker (MHC) with (w/) or without (w/o) hiPSC‐derived MNs. On the right, quantification of myotube area (μm²) per pattern with (w/) or without (w/o) hiPSC‐derived MNs. Data are presented as the mean ± SD values from at least three independent experiments in technical triplicate (at least 100 micropatterns/experiment/condition) and were analysed with Student's t‐test (ns: Not significant). (E) Representative images of immunolabelling for acetylcholine receptors clusters (AChR, Green). Scale Bar 10 μm. On the right, the quantification of the total area and the mean size of AChR clusters (μm²) per pattern as described in the supporting information method. Data are presented as the mean ± SD values from at least three independent experiments in technical triplicate (at least 100 micropatterns/condition/experiment) and were analysed with Student's t‐test (**: p < 0.005).

FIGURE 4

Immunofluorescence and ultrastructural studies reveal contacts between presynaptic and postsynaptic compartments confirmed by functional analyses. (A) Co‐immunostaining for acetylcholine receptor (AChR) and different synaptic vesicle markers: synaptophysin (SYP) and vesicular acetylcholine transporter (VAChT). Orthogonal views are represented from different planes (x/z, y/z). Scale bar: 10 μm. (B) Transmission electron micrographs of a contact between a hiPSC‐derived motoneurons (MNs) axon (obliquely cut microtubules indicated by the asterisk) and a skeletal muscle cell. Scale bar = 1 μm. The dotted square is enlarged in the right micrograph, where black arrowheads indicate synaptic vesicles in the presynaptic zone and the arrow points to plasmalemmal electron densities on both sides of the contact. Scale bar = 200 nm. (C) Representative diagram of calcium waves recorded for 100 s with (w/) or without (w/o) human‐induced pluripotent stem cell (hiPSC)‐derived MNs. ΔF/F0 representation after background correction. On the right, the quantification of the number of calcium waves per minute and per pattern in the absence or presence of hiPSC‐derived MNs. Data are presented as the mean ± SD values from three independent experiments in technical triplicate (at least 100 micropatterns/condition/experiment) and were analysed with Student's t‐test (**: p < 0.005). (D) Graphical representation of calcium waves recorded in basal condition or 4 and 8 h after the addition of a solution of 5 nM botulinum neurotoxin type‐A (BoNT/A). ΔF/F0 representation after background correction.

FIGURE 5

Synaptic defects observe at prelevel and postlevel in myotonic dystrophy type I (DM1) and muscleblind protein (MBNL) depleted coculture hybrid conditions. (A) Schematic representation of hybrid coculture systems (created with BioRender.com ). (B) Percentage of Islet1 (ISL1) ‐positive cells per pattern after 7 days of coculture. (C) Quantification of myotube area per pattern (μm²) after 7 days of coculture. (D) Quantification of neurite length (μm) per pattern after 7 days of coculture. (E) Quantification of the number of neurite branches of human‐induced pluripotent stem cell (hiPSC)‐derived motoneurons (MNs) in different hybrid coculture systems. For all the quantifications, data are represented as the mean ±SD values from at least three independent experiments in technical triplicate (at least 150 micropatterns per condition and per experiment were analysed). Data were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with control (CTRL), (**: P ≤ 0.01, ***: P ≤ 0.001, ****: P ≤ 0.0001, ns: not significant). (F) Representative acetylcholine receptor (AChR) immunolabelling (Green) images in cocultures for 7 days with MNs derived from the different hiPSC lines. Scale bar: 10 μm. On the top right, quantification of the mean size of AChR clusters per pattern (μm²) as represented as the mean ± SD values from at least three independent experiments in technical triplicate (at least 50 micropatterns/condition/experiment). Data were analysed with an ordinary one‐way ANOVA, Tukey's multiple comparisons test compared with CTRL (*: P ≤ 0.05, **: P ≤ 0.01, ns: not significant). On the bottom right, the histogram represented the proportional repartition (%) of AChR clusters by size (0–10 μm², 10–30 μm², 30 μm² and more) in different hybrid coculture systems. Data are presented as the mean ± SD values from at least three independent experiments in technical triplicate (at least 50 micropatterns/condition/experiment). Data were analysed with a two‐way ANOVA, Tukey's multiple comparisons test compared with CTRL (**: P ≤ 0.01, ***: P ≤ 0.001, ns: not significant).

FIGURE 6

Communication defect between hiPSC‐derived motoneurons (MNs) and skeletal muscle cells in myotonic dystrophy type I (DM1) and muscleblind protein (MBNL)‐depleted conditions. (A) Representative diagrams of calcium waves (time = 2 min) recorded in human primary skeletal muscle cells in different hybrid conditions (CTRL ‐control‐ in dark grey, DM1 in burgundy, DM1_3_ΔCTG in light grey and MBNL knock‐out conditions in blue). ΔF/F0 representation is indicated after background correction. (B) Quantification of the number of calcium waves recorded in human primary skeletal muscle cells per pattern and per min after 7 days of coculture. Data are presented as the mean ± SD values from at least three independent experiments in technical triplicate (at least 100 micropatterns/condition/experiment) and were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with CTRL (*: P ≤ 0.05, **: P ≤ 0.01, ***: P ≤ 0.001, ****: P ≤ 0.0001, ns: not significant).

FIGURE 7

Myotonic dystrophy type I (DM1) mutation and muscleblind protein (MBNL) depletion affect the expression of genes related to synaptic functions. (A) Hierarchical clustering of differentially expressed genes (DEG) detected in DM1 and DKO human‐induced pluripotent stem cell (hiPSC)‐derived motoneurons (MNs) compared with control (CTRL) hiPSC‐derived MNs. Gene expression is represented by a colour code ranging from green for underexpressed genes to red for overexpressed genes. (B) Schematic representation of the number of DEG in DM1 (red) and DKO (Green) hiPSC‐derived MNs. (C) Genes set enrichment analysis using EnrichR on the 129 downregulated genes in DM1 hiPSC‐derived MNs when compared with CTRL hiPSC‐derived MNs (EnrichR analysis using GO Biological process 2018 database). (D) Genes set enrichment analysis using EnrichR on the 355 downregulated genes in DKO hiPSC‐derived MNs when compared with CTRL hiPSC‐derived MNs (EnrichR analysis using GO Biological process 2018 database). (E) Genes set enrichment analysis using EnrichR on the 137 common deregulated genes between DM1 vs CTRL hiPSC‐derived MNs and DKO vs CTRL hiPSC‐derived MNs conditions (EnrichR analysis using GO Biological process 2018 database). On the right, a list of specific deregulated genes in the panel of 137 genes commonly identified between DM1 vs CTRL hiPSC‐derived MNs and DKO vs CTRL hiPSC‐derived MNs conditions. (F) Quantification by RT‐qPCR of CXCR4, DLK1 and WNT16 mRNA expression level in hiPSC‐derived MNs after normalisation with Cyclophilin A, 18S and GAPDH levels. Data are presented as the mean ± SD values from three independent experiments in technical triplicate and were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with CTRL (**: P ≤ 0.01, ***: P ≤ 0.001, ****: P ≤ 0.0001, ns: not significant).

FIGURE 8

Myotonic dystrophy type I (DM1) mutation and loss of muscleblind protein (MBNL1) and MBNL2 proteins lead to a large number of alternative splicing defects related to molecular motor activity and synaptic functions. (A) Genes set enrichment analysis using EnrichR on the 236 differentially spliced exons (DSE) in DM1 hiPSC‐derived motoneurons (MNs) when compared with CTRL hiPSC‐derived MNs (EnrichR analysis using GO Biological process 2018 database). The table on the right provides a list of specific genes differentially spliced‐in DM1 hiPSC‐derived MNs. (B) Genes set enrichment analysis using EnrichR on the 232 differentially spliced exons (DSE) in DKO hiPSC‐derived MNs when compared with CTRL hiPSC‐derived MNs (EnrichR analysis using GO Molecular Function 2018 database). The table on the right provides a list of specific genes differentially spliced‐in DKO hiPSC‐derived MNs. (C) Genes set enrichment analysis using EnrichR on the 48 common differentially spliced exons (DSE) between DM1 vs CTRL hiPSC‐derived MNs and DKO vs CTRL hiPSC‐derived MNs conditions (EnrichR analysis using GO Biological process 2018 database). On the right, list of specific misspliced genes in the panel of genes commonly identified between DM1 vs CTRL hiPSC‐derived MNs and DKO vs CTRL hiPSC‐derived MNs conditions. (D) Alternative splicing analysis for CAST exon 17 and COL13A1 exon 37 in hiPSC‐derived MNs. Data represent the mean ± SD values from three independent experiments in technical triplicate and were analysed with an ordinary one‐way analysis of variance (ANOVA), Tukey's multiple comparisons test compared with CTRL (***: P ≤ 0.001, ****: P ≤ 0.0001, ns: not significant.