Duchenne muscular dystrophy (DMD) is a progressive degenerative disease that affects cardiac, skeletal, and smooth muscle (Engel et al.,1994). It is caused by a mutation in the dystrophin gene, which encodes a large structural protein, dystrophin (Hoffman et al.,1987). There is a general concept that the lack of dystrophin causes membrane destabilization and increased calcium entry into the muscle fiber, leading to myonecrosis (Bertorini et al.,1982). In the mdx mouse model of DMD (Bulfield et al.,1984) which also completely lacks dystrophin, myonecrosis is followed by muscle regeneration (Torres and Duchen,1987; Lyons and Slater,1991).
Dystrophin and the dystroglycan complex play a key role in the anchoring of several molecules such as acetylcholine receptors (AChRs; Xu and Salpeter,1997). Changes in AChR distribution are also observed in the mdx muscle (Torres and Duchen,1987; Lyons and Slater,1991; Personius and Sawyer,2005) and parallel those seen in normal regenerated muscle fibers (Minatel et al.,2001), suggesting that these changes are a secondary consequence of muscle fiber necrosis and regeneration rather than a direct consequence of dystrophin deficiency (Lyons and Slater,1991; Minatel et al.,2003).
Dystrophin is also deficient in the spared extraocular muscles (EOMs) of mdx mice, rectus and oblique (Porter et al.,1998). However, in contrast to other muscle groups, these muscles do not show muscle fiber degeneration (Krurana et al.,1995; Porter and Baker,1996; Andrade et al.,2000). Therefore, spared EOMs are not subjected to the regeneration process and can be used to find out if the lack of dystrophin per se could cause a change in acetylcholine receptor distribution.
In the present study, we used fluorescence confocal microscopy to investigate the pattern of AChR and nerve terminal distribution in spared (rectus and oblique) and nonspared (retractor bulbi) extraocular muscles of mdx mice, and we found that the lack of dystrophin does not have any influence on the distribution of acetylcholine receptors in spared extraocular muscles of mdx mice.
MATERIALS AND METHODS
Adult mdx mice (5–6 months old) and control C57Bl/10 ScSn mice were obtained from the mouse breeding colony of the State University of Campinas and were housed on a 12-hr light/dark cycle with free access to water and standard rodent chow until the time of use. All experiments were done in accordance with the guidelines for the use of animals set forth by our Institution.
The mice were anesthetized by an intraperitoneal injection of chloral hydrate (600 μg/kg) and perfused intracardiacally with phosphate-buffered saline (PBS), followed by freshly prepared cold fixative (4% paraformaldehyde in PBS). Globes with intact extraocular muscles and associated connective tissue were obtained by orbital dissection. Individual rectus, oblique, and retractor bulbi muscles were carefully dissected from their origin to insertion. Next, the muscles (from 5 mdx and 5 C57Bl/10 mice) were placed in a Sylgard dish and washed with PBS for confocal microscopy observation of neuromuscular junction organization. Enucleated specimens from five mdx and five C57Bl/10 mice were dissected out and snap frozen with isopentane cooled in liquid nitrogen and stored at −80°C for hematoxylin and eosin and dystrophin labeling.
Frozen globes were cross-sectioned (8-μm-thick cryostat sections) transverse to the globe axis. Sections from globes were collected and mounted on coated microscope slides. Some sections were stained with hematoxylin and eosin and examined with a light microscope. The numbers of normal and regenerated muscle fibers (indicated by the presence of central cell nuclei) were counted with a hand counter.
Other sections were air-dried, hydrated for 30 min with PBS, incubated with 0.3% Triton X-100 for 10 min, and then blocked with blocking solution (1% glycine, 3% bovine serum albumin [BSA], and 0.6% Triton X-100 in PBS; Sigma) for 3 hr. The sections were incubated with dystrophin antibody (NCL-DYS1, mouse monoclonal antibody, Novocastra, 1:500) overnight at 4°C. The sections were washed with PBS and incubated with fluorescein-conjugated anti-mouse IgG (Sigma; 1:500) for 1 hr at room temperature. Sections were washed with PBS and cover-slipped with DABCO (Sigma) mounting medium for fluorescence microscopy and observed under a confocal microscope.
Control mounts for the primary antibody were incubated with fluorescein-conjugated anti-mouse IgG (Sigma; 1:500) in blocking solution instead of the primary antibody. No stained structures were seen in these controls. Approximately 200 neuromuscular junctions were examined for each group (normal and mdx).
For the study of the molecular organization of the neuromuscular junction, adult mdx mice were anesthetized with an intraperitoneal injection of chloral hydrate (0.6 g/kg) and perfused intracardiacally with PBS followed by freshly prepared cold fixative (2% formaldehyde in PBS). Right and left extraocular muscles were removed, placed in a Sylgard dish and washed with PBS. The muscles were initially incubated with rhodamine–alpha-bungarotoxin (Rh-BTX; Molecular Probes, 1 μg/ml) and then with 0.1 M glycine (Sigma) followed by a blocking solution (1% glycine, 3% BSA, and 0.6% Triton X-100 in PBS; Sigma). The primary antibody used was a mouse monoclonal anti-NF200 (Sigma, 1:500). After washing with PBS, the muscles were incubated with fluorescein-conjugated anti-mouse IgG (Sigma, 1:500) and again washed with PBS. The muscles were mounted in DABCO (mounting medium for fluorescence microscopy, Sigma) for observation as a whole-mount preparation by confocal microscopy.
The tibialis anterioris and soleus muscles of each mice had their acetylcholine receptors labeled the same way as described above for comparison with EOMs, because limb muscles show differences in terms of fiber typing, are mainly monoinnervated, and are severely affected by the lack of dystrophin.
A dual-channel Bio-Rad laser confocal system (MRC 1024UV) mounted on a Zeiss Axiovert 100 inverted microscope equipped with an Ar-Kr laser was used to observe labeled muscle fiber endplates. The 488-nm line was used to excite the fluorescein-labeled nerve terminal in muscle whole-mounts or dystrophin-labeled sections and the 568-nm line was used to excite the rhodamine-labeled receptors. Manufacturer-supplied software (Bio-Rad Acquisition and Processing) was used to control image acquisition and processing. The settings for contrast, brightness, and iris diameter were adjusted and kept unchanged during all observations of control and mdx muscles. Two Zeiss microscope objectives were used for confocal imaging: a ×40 1.4 NA water immersion objective and a ×63 1.4 NA oil immersion objective. For each endplate studied, 1- to 2-μm-thick optical sections were obtained from the bottom to the top of the endplate. Each optical section was added to a stack of images using the Bio-Rad data processing software and a single image was then built to allow observation of the whole endplate with all its branches within the same focal plane.
Based on previous observations of AChRs labeled with alpha-bungarotoxin and light microscopy (Lyons and Slater,1991; Balice-Gordon and Lichtman,1993; Marques and Santo Neto,1998; Marques et al.,2000,2005; Minatel et al.,2001,2003), junctions were qualitatively assigned according to the pattern of AChR distribution as “branched” and “islands.” The length and width of each of the AChRs clusters in an endplate were calculated. Briefly, lines following individual AChRs clusters were made with the mouse cursor (see Fig. 4, inset) and the values were automatically obtained using the Bio-Rad processing software for digital measurement. We determined that AChRs clusters with a length/width ratio equal to or less than 1.5 are classified as islands (Marques et al.,2005; Marques et al.,2007). The neuromuscular junction length and width were calculated as the maximum length of the junction in the long axis of the muscle fiber and the maximum width of the junction in the orthogonal axis, using the Bio-Rad processing software for digital measurement. Comparisons between groups were made using the Student's t-test or the χ2 test.
Histological Characteristics and Dystrophin Labeling
Mdx extraocular rectus and oblique muscles did not show any signs of muscle fiber degeneration, connective tissue accumulation, or central nuclei, characteristic of regenerated muscles (0.5% of regenerated fibers; Table 1; Fig. 1A). The mdx retractor bulbi exhibited central nucleated muscle fibers (45% of regenerated fibers; Table 1; Fig. 1B). There were no differences in the total number of fibers and fiber area between mdx and control mice for rectus, oblique, or retractor bulbi muscles (Table 1).
Table 1. Total number of fibers, percentage of fibers with central nuclei, and muscle fiber area for each extraocular and limb muscles, in control and mdx micea
Values are reported as means ± SD.
P < 0.05, significantly different from control (Student's t-test).
Dystrophin distribution was characterized by a bright outline in the sarcolemma of control muscles (Fig. 1G–I). No dystrophin labeling was observed in any dystrophic muscle studied (Fig. 1D–F). The orbital and global layers did not show any differential expression of dystrophin, with the two layers being negative in the mdx (Fig. 1D).
Neuromuscular Junction Organization
Mdx and control rectus and oblique muscles displayed a distinct innervation band in the muscle midbelly. The junctions located in this endplate zone were mainly monoinnervated (Fig. 2A,B). A few fibers (2%) showed two distinct AChR-rich areas in the same fiber, suggesting their multiple innervation (Fig. 2D). In mdx muscles, AChRs displayed the typical organization pattern described elsewhere as en plaque junctions (Khanna et al.,2003) or pretzel-like (Marques et al.,2000). The pretzel or branched pattern is classically described for the adult normal junction (Balice-Gordon and Lichtman,1993; Marques et al.,2000) and consists of elongated, smoothly fluorescent branches forming continuous channels that run in several orientations along the muscle fiber. Branches were longer than wider (Table 2). Some junctions (3–4%) showed small discrete AChR regions (Fig. 2D). The same pattern of AChR distribution was seen in control junctions (Fig. 2A; Table 2). Most of the endplates were large and oval, with a variable width of 15 μm (obliques) to 18 μm (rectus superior) and length of 30 μm (obliques) to 44 μm (rectus superior; Table 3). Nerve terminals covered the branches of the receptors with fine and continuous processes (Fig. 3A).
Table 2. Length (μm) and width (μm) of each of the AChRs branches and islands in an endplate and the percentage of junctions showing the branch and islands patterns in the spared (rectus) and nonspared (retractor bulbi) EOMs and in the limb muscle (tibialis anterior) from control and mdx micea
Values represent the mean ± SD. n = 1,000 branches/muscle.
Significantly different from respective control (P < 0.05, Student t-test and χ2).
Length/width ratio values equal to or less than 1.5 were used to classify the AChRs clusters as islands (Marques et al.,2007). AChR, acetylcholine receptor.
Table 3. Neuromuscular junction length and width obtained for each extraocular and limb muscles, in control and mdx micea
Values are reported as means ± SD. n = 200 endplates/muscle.
Junctional length (μm)
44.6 ± 13.6
46.3 ± 16.3
76.3 ± 18.7
67.8 ± 13.4
42.5 ± 7.8
31.8 ± 8.0
61.1 ± 15.6
55.0 ± 15.0
Junctional width (μm)
18.3 ± 7.2
17.0 ± 4.6
29.0 ± 6.3
39.2 ± 4.6
17.8 ± 5.4
13.6 ± 4.0
23.5 ± 5.5
29.0 ± 9.8
Controls and mdx rectus and oblique muscles also exhibited small endplates scattered along the length of the fibers proximal and distal to the central endplate zone (Fig. 3B). These junctions were classified elsewhere as en grappe (Khanna et al.,2003) and were characterized by small plaques of receptors, approximately 5.4 μm wide and 17 μm long, without much resolvable substructure and no-branched nerve terminals. No differences were seen between mdx and control muscles regarding the pattern of AChR and nerve terminal distribution at these smaller endplates.
AChRs in the mdx retractor bulbi muscles were distributed in islands (Table 2). The island pattern was characterized by round fluorescent areas with a brighter outline and a dark center (Fig. 4B–D), resembling the pattern of AChR distribution seen in limb mdx muscles (Fig. 5B,C). The islands showed approximately the same dimensions in the two axes (Table 2), and have been described before for the mdx junctions (Lyons and Slater,1991), for regenerated fibers after a crush lesion (Rich and Lichtman,1989) and after lidocaine injection (Minatel et al.,2001; Marques et al.,2005). No quantitative differences were seen in AChR pattern of distribution between mdx retractor bulbi and limb muscles (P > 0.05; Student's t-test and χ2 test; Table 2). In controls, receptors were distributed in regular and continuous branches in retractor (Fig. 4A) and tibialis anterioris (Fig. 5A). In mdx muscles, nerve terminals showed fine arborizations with bulbous enlargements that filled the center of the AChR spots (Fig. 3D). In control retractor bulbi, nerve terminals presented continuous processes that covered the AChR branches (Fig. 3C).
The histopathological analysis of EOMs showed that, despite dystrophin deficiency, mdx rectus and oblique muscles did not show central nucleated fibers, indicative of muscle fiber regeneration. Muscle regeneration was readily visible in retractor bulbi muscles of the same animal, in agreement with the literature (Pastoret and Sebille,1995; Porter and Baker,1996). Therefore, the study of mdx rectus and oblique muscles allowed us to verify whether the absence of dystrophin per se would affect acetylcholine receptor distribution at the neuromuscular junction.
The pattern of acetylcholine receptor distribution in control rectus and oblique muscles was closely similar to observations made by confocal (Khanna et al.,2003) and scanning electron microscopy (Desaki,1990), demonstrating the presence of branched and en grappe junctions. The branched junctions are typically seen in singly innervated muscle fibers and show a twitch contraction response (Spencer and Porter,1988; Khanna et al.,2003), whereas en grappe junctions are seen in multiply innervated muscle fibers with a tonic response (Khanna et al.,2003). At the present, no branched junctions were innervated by more than one nerve terminal, but a few fibers showed two distinct endplate regions.
In the dystrophic rectus and oblique muscles, the pattern of AChR organization was indistinguishable from the control muscles. No qualitative differences were seen in the pattern of receptor distribution in the branched and en grappe junctions, and no islands of receptors, usually seen in other adult dystrophic muscles, were noted (Lyons and Slater,1991; Minatel et al.,2003). Hence, these results show that the lack of dystrophin or a normal dystrophin–glycoprotein complex does not interfere with the pattern of AChR distribution at the neuromuscular junction, which is in agreement with a previous report on dystrophic neuromuscular junction development (Minatel et al.,2003).
The pattern of AChR distribution in the control retractor bulbi was similar to that described in the literature (Khanna et al.,2003). In the mdx retractor bulbi, receptors were distributed in islands and nerve terminals showed thin profiles that may represent intraterminal sprouts, which dictate the organization of AChRs into islands (Santo Neto et al.,2003). Thus, the pattern of AChR distribution in the dystrophin-deficient fiber of nonspared extraocular muscles is altered in the presence of muscle fiber regeneration, which is in agreement with previous observations in other muscles (Lyons and Slater,1991; Minatel et al.,2001). It would be interesting to verify whether limb muscles share biochemical and structural properties with EOMs, such as contraction times and myosin heavy chain components that might explain why some muscles are affected while others are not.
Extraocular muscles present a laminar organization, with an outer orbital layer and an inner global layer, which show differences of gene expression, mainly those genes related to metabolic pathways and structural elements of muscle and nerve (Budak et al.,2004). We observed that both layers were negative for dystrophin, suggesting a common mechanism of protection. Neuronal nitric oxide synthase, which is able to correct defects in neuromuscular junction in dystrophic muscles (Shiao et al.,2004) is decreased in the extraocular muscle of mdx mice (Wehling et al.,1998). Utrophin, which can be up-regulated in the dystrophic fiber, seems to not be altered in dystrophic extraocular muscles (Porter et al.,1998,2003). It would be interesting to see whether molecules of the dystrophin–glycoprotein complex related to calcium, such as calmodulin (Anderson et al.,1996; Rando,2001) can be up-regulated in dystrophic rectus and oblique muscles and could explain their protection against the lack of dystrophin.
Although the importance of dystrophin and the DGC cannot be excluded (Kong and Anderson,1999; Banks et al.,2003,2005), the present results show that the lack of dystrophin or a normal dystrophin glycoprotein complex do not have any influence on the distribution of acetylcholine receptors in spared extraocular muscles of mdx mice, which is in agreement with other studies showing a normal pattern of AChR distribution during development of dystrophic junctions (Minatel et al.,2003).
We thank Mrs. Kerstin Markendof for the English revision. H.S.N. and M.J.M. are recipients of fellowships from Conselho Nacional de Pesquisas.