The KIF1A homolog Unc-104 is important for spontaneous release, postsynaptic density maturation and perisynaptic scaffold organization

Abstract

The kinesin-3 family member KIF1A has been shown to be important for experience dependent neuroplasticity. In Drosophila, amorphic mutations in the KIF1A homolog unc-104 disrupt the formation of mature boutons. Disease associated KIF1A mutations have been associated with motor and sensory dysfunctions as well as non-syndromic intellectual disability in humans. A hypomorphic mutation in the forkhead-associated domain of Unc-104, unc-104^bris, impairs active zone maturation resulting in an increased fraction of post-synaptic glutamate receptor fields that lack the active zone scaffolding protein Bruchpilot. Here, we show that the unc-104^brismutation causes defects in synaptic transmission as manifested by reduced amplitude of both evoked and miniature excitatory junctional potentials. Structural defects observed in the postsynaptic compartment of mutant NMJs include reduced glutamate receptor field size, and altered glutamate receptor composition. In addition, we observed marked loss of postsynaptic scaffolding proteins and reduced complexity of the sub-synaptic reticulum, which could be rescued by pre- but not postsynaptic expression of unc-104. Our results highlight the importance of kinesin-3 based axonal transport in synaptic transmission and provide novel insights into the role of Unc-104 in synapse maturation.

Figure 1. Impaired synaptic transmission in unc-104^bris mutant NMJs.

(a) Representative traces of evoked release recorded from wild-type control, unc-104-OE, unc-104^bris and unc-104^bris-Rescue group. (b) Quantification of average EJP amplitude. (c) Representative traces of spontaneous release recorded from wild-type control, unc-104-OE, unc-104^bris and unc-104^bris-Rescue groups. (d–f) Quantification of mEJP amplitude (d) quantal content (e), and mEJP frequency (f) in the same groups. Unc-104^bris mutant NMJs showed significant reduction in all of the three quantified parameters, and these were rescued by pan-neuronal expression of unc-104-mcherry induced by the elav-Gal4 driver. 6–11 NMJs were quantified for each group. Martin’s correction for nonlinear summation (see material and methods) was applied for EJPs shown in (b) and for calculating quantal content. Statistical test: One-way ANOVA followed by Tukey’s Multiple Comparison Test. *P < 0.05; **P < 0.01; ***P < 0.001; n.s., P > 0.05. Error bars indicate the SEM. Genotypes: control (w¹¹¹⁸), unc-104 OE (elav-Gal4/+;UAS-unc-104-mcherry/+;), unc-104^bris (;unc-104^bris/unc-104^d11024;), rescue (elav-Gal4/+;unc-104^bris/unc-104^d11024;UAS-unc-104-mcherry/+;).

Figure 2. Impaired glutamate receptor composition at unc-104^bris mutant NMJs.

(a) Representative confocal images of control and unc-104^bris mutant NMJs marked by the presynaptic AZ protein Brp (green) and postsynaptic GluR (grey & magenta). (b) Average PSD size of unc-104^bris mutant NMJs is reduced compared to the control (control: 0.395 ± 0.019 μm²; unc-104^bris: 0.329 ± 0.013 μm². P < 0.05). (c) PSD size distribution of control and unc-104^bris mutant NMJs. There is an increased proportion of small (<0.3 μm²) and decrease of larger (>0.3 μm²), likely mature synapses in unc-104^bris mutants. (d) The sizes of PSDs apposed by presynaptic Brp are unchanged between control and unc-104^bris mutants (control: 0.407 ± 0.020 μm²; unc-104^bris: 0.412 ± 0.022 μm². P > 0.05). (e) Confocal images of neuromuscular synapses stained with Brp (blue) in control and unc-104^bris mutant larvae expressing GluRIIA-mRFP (red) and GluRIIB-GFP (green). GluRIIA was reduced in unc-104^bris mutant synapses (arrowheads). (f–h) Quantification of GluRIIA-mRFP intensity, GluRIIB-GFP intensity and GluRIIA/GluRIIB ratio at PSDs in control and unc-104^bris mutant NMJs, grouped by synapse size. (f) GluRIIA-mRFP intensity in unc-104^bris was significantly lower than control in all size groups. (g) In contrast, GluRIIB intensity in control and unc-104^bris mutants PSDs were comparable. (h) As a result, unc-104^bris mutant PSDs showed lower GluRIIA/GluRIIB ratio in all size groups, with small synapses showing stronger impairment. (i) Average size of GluRIIA-mRFP fields. (j) Average size of GluRIIB-GFP fields. For b–d and f–h, Number of NMJs analyzed: N ≥ 9 for all size groups. Statistical test: Mann-Whitney test. *P < 0.05; n.s., P > 0.05. Error bars indicate the SEM. Genotypes: control (;glurIIA-mrfp, glurIIB-gfp/+;), unc-104^bris (;unc-104^bris/unc-104^d11024;glurIIA-mrfp, glurIIB-gfp/+;).

Figure 3. Reduced intensity of postsynaptic markers in unc-104^bris mutants.

(a) A stretch of boutons stained with anti-HRP antibody (green) and antibodies against disc large (Dlg, magenta), Dorsal B (DorB, magenta) and α-Spectrin (α-Spec, magenta) showing reduced intensities at the subsynaptic reticulum. (b) Reduced Dlg expression in unc-104^bris mutant larvae. Confocal excerpts of NMJ 4 in A2 segment immuno-stained with antiserum to the scaffold protein Dlg. All genotypes were taken with the same laser intensity. The reduced expression of Dlg in unc-104^bris mutant larvae can be rescued by the pan-neuronal expression of unc-104-mcherry. Postsynaptic expression of unc-104-mcherry induced by 24B-Gal4 results in no change in the expression of Dlg compared to unc-104^bris mutants.

Figure 4. Subsynaptic reticulum development defects in unc-104^bris mutants.

(a) Electron micrographs of NMJ 4 in A2 segment. In the control group the presynaptic “T-bar” structures are clearly visible and the subsynaptic reticulum (SSR) has a compact shape. The boutons of unc-104^bris larva are considerably smaller and the SSR is less compact. Pan-neuronal expression of unc-104-mcherry rescues this defect. Postsynaptic expression of unc-104-mcherry with 24B-Gal4 does not improve the phenotype.