Deletion of the microtubule-associated protein 6 (MAP6) results in skeletal muscle dysfunction

Abstract

Background: The skeletal muscle fiber has a specific and precise intracellular organization which is at the basis of an efficient muscle contraction. Microtubules are long known to play a major role in the function and organization of many cells, but in skeletal muscle, the contribution of the microtubule cytoskeleton to the efficiency of contraction has only recently been studied. The microtubule network is dynamic and is regulated by many microtubule-associated proteins (MAPs). In the present study, the role of the MAP6 protein in skeletal muscle organization and function has been studied using the MAP6 knockout mouse line.

Methods: The presence of MAP6 transcripts and proteins was shown in mouse muscle homogenates and primary culture using RT-PCR and western blot. The in vivo evaluation of muscle force of MAP6 knockout (KO) mice was performed on anesthetized animals using electrostimulation coupled to mechanical measurement and multimodal magnetic resonance. The impact of MAP6 deletion on microtubule organization and intracellular structures was studied using immunofluorescent labeling and electron microscopy, and on calcium release for muscle contraction using Fluo-4 calcium imaging on cultured myotubes. Statistical analysis was performed using Student's t test or the Mann-Whitney test.

Results: We demonstrate the presence of MAP6 transcripts and proteins in skeletal muscle. Deletion of MAP6 results in a large number of muscle modifications: muscle weakness associated with slight muscle atrophy, alterations of microtubule network and sarcoplasmic reticulum organization, and reduction in calcium release.

Conclusion: Altogether, our results demonstrate that MAP6 is involved in skeletal muscle function. Its deletion results in alterations in skeletal muscle contraction which contribute to the global deleterious phenotype of the MAP6 KO mice. As MAP6 KO mouse line is a model for schizophrenia, our work points to a possible muscle weakness associated to some forms of schizophrenia.

Keywords: Calcium release; Microtubule-associated protein; Microtubules; Muscle contraction; Sarcoplasmic reticulum; Triad.

Fig. 1

MAP6 isoforms are present in mice skeletal muscle. a Structure of the MAP6 gene and of the major mRNA isoforms obtained by the alternative splicing of the four exons or the use of an alternative promoter in the case of MAP6-F. The dashed lines represented for each isoform, the localization of the start, and the stop codons. The localization of the antigen for the 23N antibody is represented. The localization of the primers used to amplify the four exons is represented by the arrowheads. b RT-PCR amplifications from muscle (M), brain (B), or C2C12 myotube (C) extracts, with H₂O (H) instead of cDNA as a negative control. PCR fragments at sizes corresponding to MAP6-N, E and F transcripts were amplified from brain homogenates: 443 bp for exons 1–3 (N/E), and 373 bp (major-N) and 176 bp (minor-F) for exons 2–4. In muscle homogenates and in C2C12, PCR fragments of the same size as in brain homogenates were amplified. c RT-q-PCR amplification of MAP6 transcripts in brain or muscle extracts, from three different mice, compared to β-actin. d Western blot analysis has been performed with antibody 23N on 60 ng whole brain homogenates prepared from WT or MAP6-KO animals, and 50 μg of muscle homogenate prepared from WT or MAP6-KO skeletal muscle (three mice of each genotype). In brain, five proteins are detected, the N, O, E, A, and an un-characterized 48-kDa isoform (48). The three bands specifically detected in WT skeletal muscle at 130 kDa (N-isoform), 75 kDa (E-isoform), and 42 kDa (F-isoform) are marked with arrows. The muscle-specific triadin isoform Trisk 95 (T95) has been used as a loading control (lower panel). e Western blot analysis has been performed with antibody 23N on C2C12 homogenate or 3 days WT and MAP6-KO myotube (MT) homogenates. GAPDH has been used as a loading control (lower panel)

Fig. 2

Characterization of MAP6 KO skeletal muscles. a The weight of different muscles was measured from three WT and three MAP6 KO mice and normalized to weight of WT muscles. b The cross-sectional area (CSA) was measured in type I and type II fibers from WT and MAP6 KO tibialis anterior. c Hematoxylin/eosin staining (upper panels) and NADH staining (lower panels) of WT and MAP6 KO muscles. d Quantitative RT-PCR amplification of four genes markers of denervation: AChR-γ; Musk; Myogenin; Runx1

Fig. 3

Gastrocnemius muscle mechanical performance is reduced in MAP6 KO mice. Changes in specific (a) and relative (b) twitch tension throughout a 6-min in vivo fatiguing bout of exercise performed simultaneously to bioenergetics MR acquisition. Data are means ± SEM on six WT and seven KO mice

Fig. 4

Transversal microtubules organization differs between WT and MAP6 KO fibers. A FDB muscle fibers dissociated from WT and MAP6 KO adult mice were fixed at 37 °C before labeling with anti-β-tubulin (green) and anti-RyR1 (red) antibodies. Each image represents a single confocal plane. Scale bars: 10 μm. B The microtubule network density was assessed on 40 WT and 40 KO fibers depending on its orientation: either longitudinally oriented (a) or transversally oriented (b) compared to the fiber longitudinal axis. To get more details on microtubule network organization, longitudinal lines were drawn on each cell (c), and the intensity profile was recorded for each, allowing the determination of the number of peaks per micrometer for WT and MAP6 KO cells (d), which correspond to the number of transversal microtubules per micrometer. Values are represented as means ± SEM, Student’s t test, ***p < 0.001, **p < 0.01 and ns: non-significant

Fig. 5

EM reveals a remodeling of the sarcoplasmic reticulum at the I-band in MAP6 KO fibers. a, b EM images of EDL fibers with transverse tubule (TT) staining (dark). In the insets, representative images of triads with TT pseudo-colored in green. Dotted circle in b points to an obliquely oriented TT, instead of the normal transversal TT orientation (thin black arrows). c, d EM pictures showing the organization of SR membranes at the I-band. Large arrows point to Z-lines. The free SR is colored in yellow and transverse tubules in green. e, f EM images of muscle cross sections. Empty arrows point to stacks of flat and parallel SR cisternae (one of them being enlarged in the inset). All images are representative of at least 30 fibers from 3 different animals of each genotype. Scale bars: a–b, 1 μm (inset: 0.05 μm); c–d, 0.2 μm; e–f, 0.1 μm (inset: 0.05 μm). g Representative western blots on two different mice of each genotype. h Quantitative analysis of CRC protein amounts: RyR1, alpha-1-subunit of DHPR, and SERCA in skeletal muscle homogenates. The amount of protein was normalized to GAPDH expression, the WT mean values being set to 1. n = 7 blots from three WT and three MAP6 KO mice, Mann-Whitney, ns: non-significant

Fig. 6

Calcium release is reduced in MAP6 KO myotubes compared to WT. a Direct stimulation of RyR1 by 500 μM 4CmC (black arrowhead). Fluorescence variations as a function of time are represented for WT myotubes (white dots) and MAP6 KO myotubes (black dots). Values are represented as means ± SEM for n = 150 and 151 myotubes respectively, from three different experiments, Student’s t test, ns: non-significant. b Membrane depolarization induced by addition of 140 mM KCl (black arrowhead). Fluorescence variations are presented for WT and MAP6 KO myotubes. Values are means ± SEM for n = 120 and 110 myotubes respectively, from three different experiments. Student’s t test, ***p < 0.001. c, d Maximal amplitude and area under curve for 4CmC stimulation (c) and KCl stimulation (d), with the number of analyzed-myotubes in each bar of the plot. Values are represented as means ± SEM. Student’s t test, ****p < 0.0001, **p < 0.01; ns: non-significant