Neuromuscular effects of G93A-SOD1 expression in zebrafish

Abstract

Background: Amyotrophic lateral sclerosis (ALS) is a fatal disorder involving the degeneration and loss of motor neurons. The mechanisms of motor neuron loss in ALS are unknown and there are no effective treatments. Defects in the distal axon and at the neuromuscular junction are early events in the disease course, and zebrafish provide a promising in vivo system to examine cellular mechanisms and treatments for these events in ALS pathogenesis.

Results: We demonstrate that transient genetic manipulation of zebrafish to express G93A-SOD1, a mutation associated with familial ALS, results in early defects in motor neuron outgrowth and axonal branching. This is consistent with previous reports on motor neuron axonal defects associated with familial ALS genes following knockdown or mutant protein overexpression. We also demonstrate that upregulation of growth factor signaling is capable of rescuing these early defects, validating the potential of the model for therapeutic discovery. We generated stable transgenic zebrafish lines expressing G93A-SOD1 to further characterize the consequences of G93A-SOD1 expression on neuromuscular pathology and disease progression. Behavioral monitoring reveals evidence of motor dysfunction and decreased activity in transgenic ALS zebrafish. Examination of neuromuscular and neuronal pathology throughout the disease course reveals a loss of neuromuscular junctions and alterations in motor neuron innervations patterns with disease progression. Finally, motor neuron cell loss is evident later in the disease.

Conclusions: This sequence of events reflects the stepwise mechanisms of degeneration in ALS, and provides a novel model for mechanistic discovery and therapeutic development for neuromuscular degeneration in ALS.

Figure 1

Effects of G93A-SOD1on developing zebrafish MN axons. Zebrafish embryos at the 1-4 cell stage were microinjected with wtSOD1, G93A-SOD1 and/or igf-I RNA after timed mating of adult AB strain zebrafish and underwent IHC with the SV2 antibody at 30 hpf to visualize MN axons. (A) SV2 IHC demonstrates increased branching of motor axons, represented by (*) in embryos injected with G93A-SOD1 (50 ng/μl) RNA, compared to embryos injected with wtSOD1 (50 ng/μl) RNA. 20x magnification confocal images; scale bar equals 50 μm. (B) Measurement of axon length in embryos injected with 0-100 ng/μl G93A-SOD1 RNA reveals a dose-dependent decrease in MN axon length; * P < 0.001 compared to UIC. (C) Co-injection of igf-I RNA (I; 200 ng/μl) attenuates the effects of G93A-SOD1 (G; 50 ng/μl) on axon length. No significant effects of wtSOD1 (W; 50 ng/μl) are observed, regardless of IGF-I expression. * P < 0.001 compared to UIC.

Figure 2

Transgenic ALS zebrafish. Transgenic ALS zebrafish are generated by Tol2-mediated transgenesis. (A) Constructs include a 5’ CMV promoter, human G93A-SOD1 and a 3’ EGFP C-terminal fusion protein including a SV40 late poly A tail. Microinjection of construct cDNA in the presence of Tol2 transposase RNA facilitates efficient, random integration of the transgene via flanking Tol2 sites. (B) PCR of genomic DNA extracted from 24 hpf F1 generation embryos demonstrates germline transmission of the G93A-SOD1-GFP transgene, confirming successful generation of the transgenic lines. (C) Western immunoblotting of protein lysates from 24 hpf F1 generation embryos demonstrates expression levels of the G93A-SOD1-GFP protein in three independent clutches of transgenic embryos. (D-E) Images of an F1 embryo (D; 24 hpf) and adult (E; 3 months of age) from the G21 G93A-SOD1-GFP transgenic line demonstrate stable ubiquitous expression of the fusion protein throughout the zebrafish lifespan.

Figure 3

Activity monitoring in transgenic ALS zebrafish. (A): Representative experimental layout for monitoring spontaneous swimming behavior of 20 week old control (wells 1-3) and transgenic G93A-SOD1-GFP (wells 4-6) zebrafish using the Noldus Larvae Activity Monitoring System. (B) Movement tracks extrapolated from a 1 minute trial by EthoVision for analysis of multiple different activity parameters. (C) While minimal differences are observed in the spontaneous activity levels of transgenic and control zebrafish throughout the disease course, transgenic G93A-SOD1-GFP zebrafish exhibit a consistently lower swimming velocity relative to age-matched AB controls. (D) As the disease progresses, transgenic G93A-SOD1-GFP zebrafish spend more time resting than age-matched control AB zebrafish. * P < 0.01 compared to AB control slope.

Figure 4

Neuromuscular junction integrity in transgenic ALS zebrafish. NMJs in transverse muscle sections from control AB and transgenic G93A-SOD1-GFP zebrafish were examined at multiple timepoints after αBTX staining (red) to label AChR clusters in the muscle fibers and SV2/NF IHC (green) to label MNs. Colocalization (yellow) is apparent at intact NMJs. (A-B) Representative control AB zebrafish (A) exhibit long axons with multiple synapses, whereas transgenic G93A-SOD1-GFP zebrafish (B) exhibit shorter axons and varied innervation patterns with some denervated NMJs (arrowheads). 60x oil magnification confocal images; scale bar = 20 μm. (C-D) A representative image of an individual neuron, with images included as both separated and merged fluorescent channels (C) and visualized with orthogonal XZ and YZ views of the z-series confocal image (D), validates that colocalization of SV2/NF and αBTX is representative of NMJs. 60x oil magnification confocal images; scale bar = 20 μm. (E) Quantification of SV2/NF:αBTX colocalization in 10-60 week old control AB and transgenic G93A-SOD1-GFP zebrafish. * P < 0.0001 compared to age-matched AB controls.

Figure 5

Neuromuscular innervation characterization in transgenic ALS zebrafish. Analysis of innervation patterns in 10-60 week old control AB and transgenic G93A-SOD1-GFP zebrafish. (A) Axons are scored as long (score = 1), moderately branched (scores = 2 or 3) or complex (score = 4+), as demonstrated in representative images. (B) Quantification of scoring results in 10-60 week old control AB and transgenic G93A-SOD1-GFP zebrafish. * P < 0.01 compared to age-matched AB controls.

Figure 6

MN loss in transgenic ALS zebrafish. Spinal cord cross sections from 30-60 week old zebrafish were stained with Cresyl violet Nissl (purple) to label MNs. (A) Representative low power images from control AB zebrafish is provided for orientation purposes; scale bar = 200 μm. Inset zoomed images of the spinal cord ventral horn from AB (A”) and transgenic HB9:mGFP (A’) zebrafish validate our approach to quantify MN number based on cell body size and localization to the ventral horn; scale bar = 25 μm. (B-C) Representative images from control AB (B) and transgenic G93A-SOD1-GFP (C) zebrafish at 60 weeks of age. Arrowheads denote motor neurons; 20X magnification images; scale bar = 100 μm. (D) Quantification of MN counts per ventral horn in 30, 40 and 60 week old control AB and transgenic G93A-SOD1-GFP zebrafish. * P < 0.0001 compared to age-matched AB controls.

Figure 7

Neuromuscular phenotype of transgenic ALS zebrafish. Transgenic G93A-SOD1-GFP zebrafish exhibit denervation at the NMJ around 20 weeks of age, subsequent alterations in innervation patterns at 30 weeks of age, and evidence of MN loss around 40 weeks of age which reflect a sequence of events that resembles the human disease course.