Distinct roles of the dystrophin-glycoprotein complex: α-dystrobrevin and α-syntrophin in the maintenance of the postsynaptic apparatus of the neuromuscular synapse

Abstract

α-syntrophin (α-syn) and α-dystrobrevin (α-dbn), two components of the dystrophin-glycoprotein complex, are essential for the maturation and maintenance of the neuromuscular junction (NMJ) and mice deficient in either α-syn or α-dbn exhibit similar synaptic defects. However, the functional link between these two proteins and whether they exert distinct or redundant functions in the postsynaptic organization of the NMJ remain largely unknown. We generated and analyzed the synaptic phenotype of double heterozygote (α-dbn+/-, α-syn+/-), and double homozygote knockout (α-dbn-/-; α-syn-/-) mice and examined the ability of individual molecules to restore their defects in the synaptic phenotype. We showed that in double heterozygote mice, NMJs have normal synaptic phenotypes and no signs of muscular dystrophy. However, in double knockout mice (α-dbn-/-; α-syn-/-), the synaptic phenotype (the density, the turnover and the distribution of AChRs within synaptic branches) is more severely impaired than in single α-dbn-/- or α-syn-/- mutants. Furthermore, double mutant and single α-dbn-/- mutant mice showed more severe exercise-induced fatigue and more significant reductions in grip strength than single α-syn-/- mutant and wild-type. Finally, we showed that the overexpression of the transgene α-syn-GFP in muscles of double mutant restores primarily the abnormal extensions of membrane containing AChRs that extend beyond synaptic gutters and lack synaptic folds, whereas the overexpression of α-dbn essentially restores the abnormal dispersion of patchy AChR aggregates in the crests of synaptic folds. Altogether, these data suggest that α-syn and α-dbn act in parallel pathways and exert distinct functions on the postsynaptic structural organization of NMJs.

Figure 1

Characterization and analysis of the NMJs of double heterozygote (α-dbn^+/−, α-syn^+/−) mice. α-dbn male knockout mice (α-dbn^−/−) were crossed with the α-syn female knockout mice (α-syn^−/−) to generate double heterozygote (α-dbn^+/−, α-syn^+/−) mice. (A-B) RT-PCR and Western blots of skeletal muscle homogenates (sternomastoid or gastrocnemius muscles) from young male or female bred animals (2–4 months) confirming the reduction by about half of transcript and protein expression levels of α-dbn and α-syn. Tubulin was used as a loading control. (C) Sternomastoid muscles (2–4 month-old) of wild-type and heterozygote mice were labeled with α-BTX-AlexaFluor 488 and NMJs were imaged with the confocal microscope (two representative images were shown, wildtype-type (left panel) and heterozygote (right panel)). (D) Example of muscle sections from wild-type and heterozygote mice stained with hematoxylin and eosin. Note that there were no signs of muscular dystrophy (central localization of nuclei) in heterozygote muscles. (E) Examples of NMJs on sternomastoid muscles labeled with a saturating dose of α-BTX-AlexaFluor 488 and imaged in vivo; the pseudo-color images provide a linear representation of the density of AChRs (white-yellow, high density; red-black, low density). (F) Graphs summarizing the quantification of fluorescent data from wild-type and heterozygote mice NMJs as shown in panel E. (G) Sternomastoid muscles of wild-type and age-matched heterozygote mice were bathed with a low dose of fluorescent BTX and superficial NMJs were imaged and re-imaged 3 days later. The intensity of the fluorescence was measured at each time and the total fluorescence intensity of AChR was normalized to 100% at the initial imaging. The graph shows the half-life of AChRs. Note that the half-life of AChRs in wild-type NMJs (t_1/2 of ∼10.7 ± 1.95 days, N = 20 NMJs from 3 animals) was not significantly different from the half-life of AChRs in heterozygote mice (t_1/2 ∼ 10.3 ± 2 days, N = 15 NMJs from 3 mice). WT: wild-type; D. Hetero: double heterozygotes. Data are means ± SD. Scale bars: C, E: 10 μm; D: 100 μm.

Figure 2

Characterization and analysis of the NMJs of double homozygote knock-out (α-dbn^−/−, α-syn^−/−) mice. (A) Western blots of skeletal muscle homogenates from dko confirming the absence of α-dbn and α-syn proteins. Tubulin was used as a loading control. (B) Sternomastoid muscles of young adult mice (2–3 month-old) from wild-type, α-dbn^−/−, α-syn^−/−, and dko (α-dbn^−/−; α-syn^−/−) were fixed and stained with α-BTX-AlexaFluor 488. Examples of representative images of NMJs stained with a fluorescent BTX from the above mice are shown. (C) Quantification of the length of membrane containing AChRs. Note that in NMJs from dko, the lines of AChRs extending beyond the undefined gutters were significantly longer than either α-syn or α-dbn mutant synapses. (D) Histogram showing the frequency of the abnormal AChR extensions beyond synaptic gutters. (E) Quantification of areas occupied by AChRs (synaptic size). Note that the synaptic size of single and double mutants was significantly reduced compared to wild-type, while there was no significant difference between α-syn^−/− and dko genotypes. Scale bars: 10 μm.

Figure 3

Innervation, distribution of AChRs and synaptic folds in NMJs of dko (α-dbn^−/−; α-syn^−/−). (A) Fixed sternomastoid muscles from wild-type and dko were doubly labeled with α-BTX-AlexaFluor 594 (to label AChRs) and VVA-B4-AlexaFluor 488 (marker of junctional folds). Note that in wild-type NMJs, branches display a perfect overlay between the labeling of AChR and VVA-B4 where AChRs are distributed smoothly at the crest and VVA-B4 at troughs of junctional folds. In contrast, in NMJs of dko, virtually all of the areas containing finger-like extensions of AChRs were largely devoid of VVA-B4 staining (see arrows), suggesting the absence of junctional folds. (B) Histograms show the quantification of the ratio of areas occupied by AChRs and VVA-B4. The area of α-BTX fluorescence and VVA-B4 fluorescence at each synapse was determined as described in methods, and then the ratio of the area of each synapse was calculated (****P < 0.0001). Note that in NMJs of dko significant regions of AChRs are devoid of VVA-B4 compared to single mutants and wild-type. (C) Examples of NMJs from dko and age-matched wild-type stained with α-BTX-AlexaFluor 594 to label AChRs and with antibodies against neurofilament and synaptophysin (green) to label axons. Note that in wild-type, nerve terminals (green) are in direct apposition to AChR-rich in the postsynaptic membrane, whereas in dko all the abnormal extensions of membrane containing AChRs are no longer covered by nerve terminal branches, see arrows). Data are means ± SD. Scale bars 10 μm.

Figure 4

Density and turnover rate of AChRs were further altered in NMJs of dko compared to single mutants. (A) The sternomastoid muscle of wild-type, α-dbn^−/−, α-syn^−/− and dko were bathed with α-BTX-AlexaFluor 488/594 until all AChRs were saturated and superficial NMJs were imaged. The density of AChRs was assessed as described in Methods. Shown are examples of NMJs from wild-type and mutants. The pseudo-color images provide a linear representation of the density of AChRs (white-yellow, high density; red-black, low density). (B) Graphs summarizing the quantification of fluorescent data from NMJs as shown in the panel. Note the density of AChRs was significantly reduced in dko compared to single mutants. (C) Sternomastoid muscles of wild-type, single and double mutant α-dbn and α-syn were bathed with a low dose of fluorescent BTX, and superficial NMJs were imaged and re-imaged 3 days later. The graph shows the turnover rate of AChRs. Note that the turnover rate of AChRs was significantly increased in dko compared to single mutant and wild-type. (D) Quantitative real-time PCR products of AChRα subunit from sternomastoid muscles of all genotypes. (E) Quantitative real-time PCR products of AChRα subunit from synaptic and extra-synaptic regions of dko and wild-type sternomastoid muscles. Note that there was no significant difference in AChRα subunit transcript levels between all genotypes (at least 3 mice were used per experiment. Data are means ± SEM. Scale bars: 10 μm.

Figure 5

Synaptic strength and force generation are reduced in dko and single α-dbn^−/−, while the distribution of DGC and basal lamina components remain unaffected. (A) Sections of sternomastoid muscles from wild-type and dko were stained with antibodies against DGC complex and laminin gamma 1 (γ1) and imaged with a confocal microscope. No changes in the localization of either β-dystroglycan, dystrophin, or laminin γ1 around the membrane of the muscle cell of dko compared to wild-type. (B) Wild-type, single and dko mice were forced to run on a treadmill and the time to exhaustion of each animal was recorded as shown in Histogram B. Note that the time to exhaustion of dko and α-dbn^−/− mice was significantly shorter than α-syn^−/− and wild-type mice. (C) Histogram showing the forelimb grip strength. Note that there was no significant difference in the grip strength between dko and α-dbn^−/− mutant mice, but it was significantly reduced compared to α-syn^−/− and wild-type mice. (D) Hematoxylin and eosin on sections of sternomastoid muscles from adult dko, single α-dbn^−/−, α-syn^−/− and wild-type mice. (E) Histogram showing the quantification of muscles containing central nuclei (signs of muscle regeneration). Data are means ± SD. Scale bars: 10 μm (A); and 100 μm (D).

Figure 6

α-dbn or α-syn partially rescues the density of AChRs at NMJs of dko. (A) Example of an NMJ from the sternomastoid muscle of young adult α-dbn-null mice electroporated with α-dbn-GFP. Note that the α-dbn-GFP construct restores the abnormal distribution of AChRs to normal when compared to non-electroporated synapses and wild-type. (B) Example of an NMJ from the sternomastoid muscle of young adult α-syn-null mice electroporated with α-syn-GFP. Note that the expression of α-syn-GFP restores the abnormal pattern of AChR distribution to normal when compared to wild-type synapse and non-electroporated mutant NMJs. (C) Examples of NMJs from the sternomastoid muscle of dko electroporated with either α-syn-GFP, α-dbn-GFP, or a pair of constructs containing the α-dbn N- (VN) and α-syn C- (VC) terminal fragments of Venus protein (α-syn-VC and α-dbn-VN) and labeled with a saturating dose of α-BTX-AlexaFluor 594 as described previously (30). (D) Quantification of fluorescent AChRs data from NMJs as shown in C. Data are means ± SD. Scale bars 10 μm.

Figure 7

α-syn and α-dbn perform distinct functions in the maintenance and distribution of AChR clusters in the postsynaptic membrane. The sternomastoid muscle of young adult male and female dko mice were electroporated with either α-dbn1-GFP or α-syn-GFP and 10 to 15 days later, the sternomastoid was bathed with α-BTX-AlexaFluor594 to label AChRs. Electroporated and non-electroporated neighboring synapses were then imaged with a confocal microscope. (A) Example of wild-type (top left panel) and non-electroporated dko (top right panel) NMJs. Images in the middle panels represent an example of electroporated NMJ expressing α-syn-GFP and labeled with α-BTX-AlexaFluor 594. Images in the bottom panels represent an example of electroporated NMJ expressing α-dbn1-GFP and labeled with α-BTX-AlexaFluor 594 (arrows showed the presence of long extensions beyond synaptic edges). (B) Graph showing the quantification of the average lengths of finger-like extensions (streaks) in electroporated and non-electroporated NMJs with either α-syn-GFP, α-dbn-GFP, or α-syn-VC and α-dbn-VN. (C) Graph showing the quantification of the frequency of abnormal streaks per synapse at electroporated muscles with either α-syn-GFP, α-dbn-GFP, or α-syn-VC and α-dbn-VN and non-electroporated NMJs. (D). Graph showing the quantification of abnormal patchy micro-clusters of AChR finger-like (hot spots) in electroporated and non-electroporated dko muscles with the above constructs). (E) A schematic representation showing the potential roles of α-dbn and α-syn in the structural organization of the postsynaptic apparatus. In the absence of both α-dbn and α-syn in the muscle, many postsynaptic regions including the synaptic folds and AChR density were highly simplified (reduced). However, when α-dbn was added AChR anchoring/localization at the crests of synaptic folds was significantly improved, while the addition of α-syn essentially restores junctional folds. Data are means ± SD. Scale bars 10 μm.