Drosophila Nesprin-1 controls glutamate receptor density at neuromuscular junctions

Abstract

Nesprin-1 is a core component of a protein complex connecting nuclei to cytoskeleton termed LINC (linker of nucleoskeleton and cytoskeleton). Nesprin-1 is anchored to the nuclear envelope by its C-terminal KASH domain, the disruption of which has been associated with neuronal and neuromuscular pathologies, including autosomal recessive cerebellar ataxia and Emery-Dreifuss muscular dystrophy. Here, we describe a new and unexpected role of Drosophila Nesprin-1, Msp-300, in neuromuscular junction. We show that larvae carrying a deletion of Msp-300 KASH domain (Msp-300 (∆KASH) ) present a locomotion defect suggestive of a myasthenia, and demonstrate the importance of muscle Msp-300 for this phenotype, using tissue-specific RNAi knock-down. We show that Msp-300 (∆KASH) mutants display abnormal neurotransmission at the larval neuromuscular junction, as well as an imbalance in postsynaptic glutamate receptor composition with a decreased percentage of GluRIIA-containing receptors. We could rescue Msp-300 (∆KASH) locomotion phenotypes by GluRIIA overexpression, suggesting that the locomotion impairment associated with the KASH domain deletion is due to a reduction in junctional GluRIIA. In summary, we found that Msp-300 controls GluRIIA density at the neuromuscular junction. Our results suggest that Drosophila is a valuable model for further deciphering how Nesprin-1 and LINC disruption may lead to neuronal and neuromuscular pathologies.

Fig. 1

Msp-300 subcellular localization and contribution to nuclei anchoring. a Schematic representation of the LINC complex. b–d Confocal images of WT and Msp-300 ^∆KASH third-instar wandering larvae showing Msp-300 subcellular localization (in grey). b Msp-300 localizes at Z-bands (revealed by Kettin, D-Titin) in a punctate pattern in both WT (left) and Msp-300 ^∆KASH (right) larvae. Top panels show Msp-300 staining, middle panels Kettin staining, and lower panels the overlay with Msp-300 in green and Kettin in magenta. c, d Perinuclear localization of Msp-300 in muscle 4. The nuclear envelope is labeled using an anti LaminA antibody (in green). The lower panels show Msp-300 staining alone. c Msp-300 is found apposed to the nuclear envelope. In Msp-300 ^∆KASH, Msp-300 retracts from the nucleus, forming a ring disconnected from the nuclear envelope. d In WT larvae (left panels), Msp-300 localizes in puncta arranged in strings and forming a light net at the surface of the fiber as shown on the middle panel. The lower panel is an enhanced image of the nucleus surface revealing the homogeneous web of Msp-300 staining. In Msp-300 ^∆KASH (right panels), surface localization of Msp-300 is less regular and occasional spikes of Msp-300 can be observed (in 34 % of nuclei, n = 47). e Muscles 6 and 7 of WT and Msp-300 ^∆KASH larvae showing muscle (in red) and NMJ (in green) outlines and nuclei (stained with anti LaminA, in grey). Nuclei are evenly spaced in WT larvae but clustered in Msp-300 ^∆KASH. f Clustering quantification in fibers 4, 6, and 7, comparing WT and Msp-300 ^∆KASH conditions and showing the percentage of total nuclei found isolated or in groups of two to more than seven nuclei. Scale bars in b, c and d: 10 µm, in e: 100 µm

Fig. 2

Msp-300 ^∆KASH larvae show altered locomotion activity. a Box plot representation of the global displacement of 48 h AEL old larvae following 2 min of wandering relative to larval size (WT, n = 13, Msp-300 ^∆KASH, n = 16). b–g Fine locomotion analysis of WT and Msp-300 ^∆KASH larvae obtained by time-lapse imaging of 48 h AEL old larvae. b Contraction amplitude, i.e., ratio of the shortest to the longest size of the larva, directly reflecting longitudinal muscles contraction (WT, n = 16, Msp-300 ^∆KASH, n = 16). c Instant speed of the larva, extracted from pure forward movement sequences (straight) relative to larval size (WT, n = 16, Msp-300 ^∆KASH, n = 19). Msp-300 ^∆KASH values in b and c are not significantly different from WT (respectively, p = 0.13 and p = 0.60). (a–c) The values were normalized to the median value obtained with WT larvae. (d–e) Larval tracks extracted from 1 min long locomotion sequences representative of the WT and Msp-300 ^∆KASH behaviors. Locomotion was imaged by time-lapse imaging at a rate of 1 frame/150 ms. Larvae outlines are colored by groups of 40 images (6 s) and superposed to show progression of the locomotion. A star marks the position of the first image of each group. Different behaviors can be observed on these tracks, including pauses shown by the superposition of stars (in e). Scale bar 1 mm. f–g Quantification of the different behaviors of WT (grey, n = 16) and Msp-300 ^∆KASH (green, n = 19) larvae, a dark line shows the median value for each category. f Total time spent adopting a given behavior represented as the percentage of the total locomotion sequence’s length. A given larva will be represented by one point in each behavioral category. WT and Msp-300 ^∆KASH larvae significantly differ in the percentage of time spent in the straight and pause behaviors. g Duration of behavior sequences in seconds for each larva. Each larva will be represented by several points for each behavioral category. Msp-300 ^∆KASH larvae are characterized by significantly shorter straight and longer pause sequences than WT larvae. **p < 10^–3, Wilcoxon test

Fig. 3

Msp-300 KASH domain deletion results in postsynaptic alteration of NMJ function. a Comparison of the global displacement relative to larval size following 2 min of wandering of 48 h AEL old control larvae (;UAS-Dicer2;, n = 7) and 48 h AEL old larvae with either neuron (Elav-Gal4; UAS-Dicer2;, n = 14) or muscle (24B-Gal4, n = 16 and UAS-Dicer2; 24B-Gal4, n = 14)-specific RNAi knock-down of Msp-300. For the ease of comparison, the median value obtained for control larvae was adjusted to 100. While neuron-specific knock-down of Msp-300 does not alter larval crawling capacities, we observe a marked decrease in the distance/larval size for the 24B-Gal4 progeny. This decrease is enhanced by the potentiation of RNAi upon Dicer2 expression (UAS-Dicer2; 24B-Gal4 progeny), hence showing the specificity of the RNAi knock-down (78 and 22 % of control distance/size, respectively, Wilcoxon test, *p = 0.0012, **p = 0.0003). b–f Voltage clamp recordings from muscle 6 of abdominal segment 3 of WT and Msp-300 ^∆KASH. Representative traces of spontaneous neurotransmitter release (b) and evoked synaptic transmission (c) recorded from WT and Msp-300 ^∆KASH larvae. d The amplitude of mEJCs is not different in Msp-300 ^∆KASH and WT (Wilcoxon test, p = 0.79) while eEJCs amplitude (e) and quantal content (f, eEJC/mEJC) are significantly reduced in Msp-300 ^∆KASH when compared to WT (Wilcoxon test, p = 0.0028 and p = 0.0115, respectively)

Fig. 4

Composition in glutamate receptors is altered in Msp-300 ^∆KASH larvae. a Fiber 4 of WT and Msp-300 ^∆KASH wandering third instar were labeled with GluRIIC (top panels and in magenta in the third) and GluRIIA (second panels, green in the third) to quantify the proportion of GluRIIA-containing receptors at the NMJ. Bottom panels show IIC dots colocalizing with IIA. b Representative measures of GluRIIA density in WT and Msp-300 ^∆KASH showing fluorescence density of GluRIIA staining relative to WT median value. We observe a significant decrease in GluRIIA density in Msp-300 ^∆KASH (med = 65.8 % of control, n = 14) compared to WT (n = 10; Wilcoxon test, p = 0.01) (experiment repeated four times). c Measure of GluRIIC density in WT and Msp-300 ^∆KASH showing GluRIIC fluorescence density relative to WT median value. We did not observe any significant difference between WT (n = 32) and Msp-300 ^∆KASH (n = 34). d Representative quantification of GluRIIA-containing receptors given by Mander’s colocalization coefficient of GluRIIC and GluRIIA and showing a significant decrease in the proportion of A-type GluR at Msp-300 ^∆KASH NMJ (59 %, n = 8 vs. 71 % in WT, n = 12; Wilcoxon test, p = 0.005). Scale bar 10 µm

Fig. 5

GluRIIA junctional decrease in Msp-300 ^∆KASH larvae results in locomotion impairment. a, c, d Distance run in 2 min by 48 h AEL old larvae relative to larval size. a Comparison of global locomotion of WT (n = 13), Msp-300 ^∆KASH (n = 15), and gluRIIA[AD9]/+ (n = 16) larvae. Larvae carrying a single copy of gluRIIA gene behave as Msp-300 ^∆KASH larvae. b Comparison of fluorescence density of GluRIIA staining in WT, Msp-300 ^∆KASH, Msp-300 ^∆KASH, MHC-GluRIIA::myc and MHC-GluRIIA::myc. Muscle overexpression of GluRIIA rescues GluRIIA fluorescence density of Msp-300 ^∆KASH larvae to WT levels. GluRIIA fluorescence density of MHC-GluRIIA::myc do not differ from WT levels, suggesting that GluRIIA is not limiting in WT conditions. c Global locomotion of Msp-300 ^∆KASH larvae over-expressing gluRIIA specifically in muscles (Msp-300 ^∆KASH, MHC-GluRIIA::myc, n = 22) compared with distance traveled by WT (n = 22) and Msp-300 ^∆KASH (n = 22) larvae. Distance values of Msp-300 ^∆KASH, MHC-GluRIIA::myc larvae significantly differ from values of WT (Wilcoxon test, p = 0.005) or Msp-300 ^∆KASH larvae (p = 0.02). Sixty percent of the over-expressing larvae perform better than Msp-300 ^∆KASH larvae, therefore showing a significant rescue of their locomotion abilities. d WT (n = 24) and GluRIIA overexpressing larvae (MHC-GluRIIA::myc, n = 26) perform identically in global locomotion tests, showing that GluRIIA is not limiting in WT conditions (Wilcoxon test, p = 0.1932)

Fig. 6

Dorsal and Cactus sub-synaptic localizations are not affected by Msp-300 KASH domain deletion. Fiber 4 of WT (left panels) and Msp-300 ^∆KASH (right panels) wandering third-instar larvae labeled with Dorsal (a) and Cactus (c) together with HRP to show neuromuscular junction extent. Representative measures of Dorsal (b) and Cactus (d) fluorescence densities in WT and Msp-300 ^∆KASH relative to WT median value. We did not observe any significant difference between WT and Msp-300 ^∆KASH larvae neither in Dorsal (WT, n = 8, Msp-300 ^∆KASH, n = 9, Wilcoxon test, p = 0.32), or in Cactus fluorescence densities (WT, n = 8, Msp-300 ^∆KASH, n = 9, Wilcoxon test, p = 0.96). Scale bar 10 µm. The experiment was repeated three times for each staining

Fig. 7

ens ^swo mutants present locomotion defects together with altered Msp-300 localization but no nuclei clustering. a Box plot representation of the global displacement of 48 h AEL old WT and ens ^swo larvae following 2 min of wandering relative to larval size. Two independent tests including eight larvae per genotype were performed. In both tests, ens ^swo larvae covered a distance corresponding to <60 % of the distance covered by WT larvae (p < 10^–3, Wilcoxon test). b Clustering quantification in muscles 4, 6, and 7 of ens ^swo third instar larvae and showing the percentage of total nuclei found isolated or in groups of two to more than seven nuclei. We analyzed 83 fibers from 11 larvae and only found four of the 1,150 nuclei counted in clusters of two (<2 % of the muscle four nuclei counted). Thus, there is no nuclei clustering in ens ^swo third-instar larvae. c–e Subcellular localization of Msp-300 (grey) in WT (c) and ens ^swo (d, e) muscle 4 of third instar larvae. While Z-band localization of Msp-300 appears normal in ens ^swo mutants, Msp-300 perinuclear organization is altered. In all fibers, we observe a disorganized Msp-300 staining above the sarcomeres in the region surrounding the nuclei (labeled by LaminA in green, compare c and e). In addition, we observed in some fibers a gap in Msp-300 staining around the nuclei, which is very similar to Msp-300 organization observed in Msp-300 ^∆KASH mutants (compare c and d, see Fig. 1c). Scale bar 20 µm