The Mouse Levator Auris Longus Muscle: An Amenable Model System to Study the Role of Postsynaptic Proteins to the Maintenance and Regeneration of the Neuromuscular Synapse

Abstract

The neuromuscular junction (NMJ) is the peripheral synapse that controls the coordinated movement of many organisms. The NMJ is also an archetypical model to study synaptic morphology and function. As the NMJ is the primary target of neuromuscular diseases and traumatic injuries, the establishment of suitable models to study the contribution of specific postsynaptic muscle-derived proteins on NMJ maintenance and regeneration is a permanent need. Considering the unique experimental advantages of the levator auris longus (LAL) muscle, here we present a method allowing for efficient electroporation-mediated gene transfer and subsequent detailed studies of the morphology and function of the NMJ and muscle fibers. Also, we have standardized efficient facial nerve injury protocols to analyze LAL muscle NMJ degeneration and regeneration. Our results show that the expression of a control fluorescent protein does not alter either the muscle structural organization, the apposition of the pre- and post-synaptic domains, or the functional neurotransmission parameters of the LAL muscle NMJs; in turn, the overexpression of MuSK, a major regulator of postsynaptic assembly, induces the formation of ectopic acetylcholine receptor clusters. Our NMJ denervation experiments showed complete reinnervation of LAL muscle NMJs four weeks after facial nerve injury. Together, these experimental strategies in the LAL muscle constitute effective methods to combine protein expression with accurate analyses at the levels of structure, function, and regeneration of the NMJ.

Keywords: electroporation; neuromuscular junction; postsynaptic; presynaptic; regeneration; skeletal muscle.

Figure 1

Efficient electroporation-mediated gene transfer of Levator auris longus (LAL) muscles. LAL muscles from adult mice were electroporated in vivo with a plasmid coding for the tdTomato protein. Plasmid DNA was injected underneath the LAL muscle fascia and gold needle-type electrodes were positioned above the muscle to deliver five pulses of 100 V/cm² of 20 ms of duration at 1 Hz (A). After 21 days, dissected whole-mounted muscles (B) were transversally sectioned and stained with Alexa488-WGA (green) and DAPI (blue) (C) to label the muscle membrane and nuclei, respectively. (D) Higher magnification images were used to quantify the efficiency of the procedure [as the percentage of tdTomato-expressing fibers from total fibers quantified based on wheat germ agglutinin (WGA) staining] (E) and the presence of central nuclei (arrowheads), as a parameter of muscle fiber damage/regeneration (E). (F) Transversal cryosections stained with Hematoxylin/Chromotrope revealed no significant alterations in muscle fiber histology or mononuclear cell infiltration. (G) NADH-TR histochemical activity detection was used to analyze non-oxidative (light and middle blue) and oxidative (dark blue) fibers. The proportion of these fiber types was quantified and expressed as a percentage of total fibers in the region of interest (H). Also, the cross-sectional area (CSA) of oxidative (I) and non-oxidative (J) fibers was determined in >100 fibers per type in each Hemi-LAL. The results represent the mean ± SEM of N: three mice per group (control and electroporated). Scale bar 5 mm (B), 200 μm (C), 50 μm (D), 50 μm (F,G). p > 0.05, t-test; n.s.= non-significant.

Figure 2

The in vivo electroporation procedure in the LAL muscle does not affect neuromuscular junction (NMJ) innervation. LAL muscles from control adult mice and those expressing the tdTomato protein for 21 days were dissected and subjected to immunohistochemistry with the 2H3 (neurofilament) plus SV2 (synaptic vesicles) antibodies to reveal presynaptic motor terminals, along with Alexa647-BTX (white pseudocolor) to stain postsynaptic densities. (A) Low magnification images of whole-mount preparations show that the NMJ profile from the R5 region of right Hemi LAL muscles is maintained in tdTomato-expressing fibers, as compared to controls. (B,C) Higher magnification confocal images show that terminal motor axon branches contact postsynaptic apparatuses in control and electroporated muscle fibers. (D) Quantification of the apposition of postsynaptic acetylcholine receptor (AChR) pretzels by presynaptic motor axons. The plots correspond to >45 NMJs per mice. The bars represent the mean ± SEM of N: three mice per group (control and electroporated). Scale bar 200 μm (A), 5 μm (B,C). p > 0.05, t-test; n.s.= non-significant.

Figure 3

Neuromuscular transmission is not altered after electroporation-mediated gene transfer of the LAL muscle. For functional studies, LAL muscles overexpressing the tdTomato protein at P14 were analyzed through electrophysiological intracellular recording at P21. Non-electroporated LAL muscles from age-matched mice were used as controls. After blocking muscle contraction, stimulation trains of 100 Hz during 1 s (A) showed no changes in synaptic plasticity, evidenced by paired-pulse facilitation (PPF) (B) and depression index (C) quantification in control and electroporated fibers. Plots represent the average ± SEM of N: 3, n: 24 (control and electroporated fibers). (D) Representative end plate potential (EPP) and miniature EPP (mEPP) traces after 0.5 Hz stimuli of NMJs from control and tdTomato-expressing fibers. Amplitudes of EPPs (E), mEPPs (F), and the quantal content (G) of control and electroporated fibers were quantified. Plots represent the average ± SEM of N: 3, n: 24 (control and electroporated fibers). p > 0.05, t-test; n.s.= non-significant.

Figure 4

Electroporation-mediated gene transfer of tdTomato does not alter NMJ postsynaptic maturation. LAL muscles from control adult mice and those expressing the tdTomato protein for 10 and 21 days were dissected and stained with Alexa488-BTX. (A) AChR pretzels on the surface of tdTomato-expressing fibers were similar to those on control non-electroporated fibers (A). To analyze NMJ maturation, electroporated LAL muscles at P21 were analyzed at P42. Non-electroporated age-matched mice were used as controls. AChR clusters were categorized into plaques (P), perforated plaques (PP), immature pretzels (IP), mature pretzels (MP), and fragmented pretzels (FP; B, upper panel) and their relative abundance were plotted (B, lower panel). Postsynaptic area (C) and perimeter (D) were also determined. Plots represent the average ± SEM of N: 3, n: 130 (Control), and N:3, n: 149 (tdTomato-expressing fibers). Scale bar 50 μm (A), 25 μm (B). p > 0.05, t-test; n.s.= non-significant.

Figure 5

Electroporation-mediated overexpression of the MuSK receptor results in ectopic AChR aggregation. LAL muscles were transversally sectioned and stained with WGA conjugated to Alexa488 (green) to label the muscle membrane at different times after electroporation (A). (B) Quantification of electroporation efficiency at 3, 7, and 14 days after electroporation. To analyze the efficiency of our procedure to modulate postsynaptic organization at the NMJ, LAL muscles from adult mice were co-electroporated with plasmids coding for tdTomato and rMuSK-myc to overexpress a myc-tagged version of the rat MuSK receptor and tdTomato in a 5:1 (rMuSK-myc:tdTomato) proportion. Age-matched mice subjected to tdTomato electroporation only were used as controls. (C) Western blot using an anti-myc antibody was performed on protein extracts from HEK293 cells and LAL muscles electroporated with tdTomato and co-electroporated with tdTomato and rMuSK-myc. After 21 days, LAL muscles were dissected and subjected to immunohistochemistry with the 2H3 (neurofilament) plus SV2 (synaptic vesicles) antibodies to reveal presynaptic motor terminals (white pseudocolor) along with Alexa488-BTX to stain postsynaptic densities (D). (E) LAL muscles were also electroporated with the MuSK-EGFP plasmid, which contains the full-length mouse MuSK coding sequence. We used a plasmid to express EGFP as control. After 14 days, LAL muscles were dissected and subjected to immunohistochemistry with the 2H3 (neurofilament) plus SV2 (synaptic vesicles) antibodies to reveal presynaptic motor terminals (red) along with Alexa647-BTX to stain postsynaptic densities (white pseudocolor). Images at the top of the right column are magnified images of the left panels. Images at the bottom of the right column show transversal cryosections of the same LAL muscles in synaptic (Syn) and extrasynaptic (ExSyn) regions of the muscle fiber. Arrows in panels (D,E) show AChR clusters in extrasynaptic regions of MuSK-overexpressing LAL fibers. Scale bar 100 μm (A), 200 μm (D), 50 μm (E, left and right top panels), 10 μm (right bottom panels).

Figure 6

LAL muscle after degenerative and regenerative nerve damage. To study NMJ regeneration in the LAL muscle, the posterior auricular branch of the facial nerve was subjected to two protocols of injury: a portion of approximately 4 mm was transected to induce NMJ degeneration, whereas the facial nerve was crushed for 30 s to accomplish NMJ regeneration (A). To analyze NMJ behavior after nerve injury, LAL muscles in the different treatments were dissected 7 (upper panels) and 30 days (lower panels) after facial nerve damage. (B) AChRs were labeled with α-bungarotoxin (αBTX; green), motor axons were stained with antibodies against neurofilaments, 2H3, and synaptic vesicle proteins, SV2 (red), and Schwann cells were stained with the anti S100 antibody (white). The areas of the presynaptic motor terminal (C) and the postsynaptic AChR-rich domain (D) were measured. Quantification of the apposition of postsynaptic AChR pretzels by presynaptic motor axons (E). Plots represent the average ± SEM of N: 4, n: 102 (Control, black bars), N: 3, n: 116 (7 days after crush, blue bars), and N: 3, n: 105 (30 days after crush, orange bars; **p < 0.01; ***p < 0.001, ****p < 0.0001, one-way ANOVA). Scale bar 25 μm (B).